CINXE.COM
ODSC West Schedule 2024 | Open Data Science Conference
<!doctype html> <!-- paulirish.com/2008/conditional-stylesheets-vs-css-hacks-answer-neither/ --> <!--[if lt IE 7]><html class="no-js lt-ie9 lt-ie8 lt-ie7" lang="en"> <![endif]--> <!--[if IE 7]><html class="no-js lt-ie9 lt-ie8" lang="en"> <![endif]--> <!--[if IE 8]><html class="no-js lt-ie9" lang="en"> <![endif]--> <!-- Consider adding a manifest.appcache: h5bp.com/d/Offline --> <!--[if gt IE 8]><!--><html class="no-js" lang="en-US"> <!--<![endif]--><head><meta charset="utf-8"><meta http-equiv="X-UA-Compatible" content="IE=edge,chrome=1" ><meta name="viewport" content="user-scalable=no, width=device-width, initial-scale=1, maximum-scale=1" /><meta name="apple-mobile-web-app-capable" content="yes" /><link href="//maxcdn.bootstrapcdn.com/font-awesome/4.7.0/css/font-awesome.min.css" rel="stylesheet"> <style> #icon_wrapper .fuse_social_icons_links .fa { font-family: 'FontAwesome' !important; } .fb-awesome-social::before { content: "\f09a" !important; } #icon_wrapper { position: fixed; top: 25%; right: 0px; z-index: 99999; } .awesome-social { margin-top:2px; color: white !important; text-align: center !important; line-height: 25px !important; width: 24px !important; height: 24px !important; font-size: 1em !important; -moz-transition: width 0.5s, height 0.5s, -webkit-transform 0.5s; /* For Safari 3.1 to 6.0 */ -webkit-transition: width 0.5s, height 0.5s, -webkit-transform 0.5s; /* For Safari 3.1 to 6.0 */ transition: width 0.5s, height 0.5s, transform 0.5s; } .awesome-social:hover { -webkit-transform: rotate(360deg); /* Chrome, Safari, Opera */ transform: rotate(deg); -moz-transform: rotate(360deg); /* Chrome, Safari, Opera */ -ms-transform: rotate(360deg); /* Chrome, Safari, Opera */ } .fuse_social_icons_links { outline:0 !important; } .fuse_social_icons_links:hover{ text-decoration:none !important; } .fb-awesome-social { background: #3b5998; } .tw-awesome-social { background:#00aced; } .rss-awesome-social { background:#FA9B39; } .linkedin-awesome-social { background:#007bb6; } .youtube-awesome-social { background:#bb0000; } .flickr-awesome-social { background: #ff0084; } .pinterest-awesome-social { background:#cb2027; } .stumbleupon-awesome-social { background:#f74425 ; } .google-plus-awesome-social { background:#f74425 ; } .instagram-awesome-social { background: -moz-linear-gradient(45deg, #f09433 0%, #e6683c 25%, #dc2743 50%, #cc2366 75%, #bc1888 100%); background: -webkit-linear-gradient(45deg, #f09433 0%,#e6683c 25%,#dc2743 50%,#cc2366 75%,#bc1888 100%); background: linear-gradient(45deg, #f09433 0%,#e6683c 25%,#dc2743 50%,#cc2366 75%,#bc1888 100%); filter: progid:DXImageTransform.Microsoft.gradient( startColorstr='#f09433', endColorstr='#bc1888',GradientType=1 ); } .tumblr-awesome-social { background: #32506d ; } .vine-awesome-social { background: #00bf8f ; } .vk-awesome-social { background: #45668e ; } .soundcloud-awesome-social { background: #ff3300 ; } .reddit-awesome-social{ background: #ff4500 ; } .stack-awesome-social{ background: #fe7a15 ; } .behance-awesome-social{ background: #1769ff ; } .github-awesome-social{ background: #999999 ; } .envelope-awesome-social{ background: #ccc ; } /* Mobile */ /* Custom Background */ .awesome-social { background:#323840 !important; } .awesome-social{ -webkit-transition-property:color, text; -webkit-transition-duration: 0.25s, 0.25s; -webkit-transition-timing-function: linear, ease-in; -moz-transition-property:color, text; -moz-transition-duration:0.25s; -moz-transition-timing-function: linear, ease-in; -o-transition-property:color, text; -o-transition-duration:0.25s; -o-transition-timing-function: linear, ease-in; } .fb-awesome-social:hover { color: #3b5998 !important; } .tw-awesome-social:hover { color:#00aced !important; } .rss-awesome-social:hover { color:#FA9B39 !important; } .linkedin-awesome-social:hover { color:#007bb6 !important; } .youtube-awesome-social:hover { color:#bb0000 !important; } .flickr-awesome-social:hover { color: #ff0084 !important; } .pinterest-awesome-social:hover { color:#cb2027 !important; } .stumbleupon-awesome-social:hover { color:#f74425 !important; } .google-plus-awesome-social:hover { color:#f74425 !important; } .instagram-awesome-social:hover { color:#517fa4 !important; } .tumblr-awesome-social:hover { color: #32506d !important; } .vine-awesome-social:hover { color: #00bf8f !important; } .vk-awesome-social:hover { color: #45668e !important; } .soundcloud-awesome-social:hover { color: #ff3300 !important; } .reddit-awesome-social:hover{ color: #ff4500 !important; } .stack-awesome-social:hover{ color: #fe7a15 !important; } .behance-awesome-social:hover{ color: #1769ff !important; } .github-awesome-social:hover{ color: #999999 !important; } </style><meta name='robots' content='index, follow, max-image-preview:large, max-snippet:-1, max-video-preview:-1' /> <!-- Google Tag Manager for WordPress by gtm4wp.com --> <script data-cfasync="false" data-pagespeed-no-defer> var gtm4wp_datalayer_name = "dataLayer"; var dataLayer = dataLayer || []; </script> <!-- End Google Tag Manager for WordPress by gtm4wp.com --> <!-- This site is optimized with the Yoast SEO Premium plugin v23.6 (Yoast SEO v23.8) - https://yoast.com/wordpress/plugins/seo/ --><title>ODSC West Schedule 2024 | Open Data Science Conference</title><link rel="stylesheet" id="siteground-optimizer-combined-css-518a7a5fa74639197a615aa89ad5b0db" href="https://odsc.com/wp-content/uploads/siteground-optimizer-assets/siteground-optimizer-combined-css-518a7a5fa74639197a615aa89ad5b0db.css" media="all" /><link rel="preload" href="https://odsc.com/wp-content/uploads/siteground-optimizer-assets/siteground-optimizer-combined-css-518a7a5fa74639197a615aa89ad5b0db.css" as="style"><meta name="description" content="ODSC West 2022 West Preliminary Session Schedule" /><meta property="og:locale" content="en_US" /><meta property="og:type" content="article" /><meta property="og:title" content="ODSC West Schedule 2024" /><meta property="og:description" content="ODSC West 2022 West Preliminary Session Schedule" /><meta property="og:url" content="https://odsc.com/california/odsc-west-schedule-2023/" /><meta property="og:site_name" content="Open Data Science Conference" /><meta property="article:publisher" content="https://www.facebook.com/OPENDATASCI" /><meta property="article:modified_time" content="2024-10-31T22:07:23+00:00" /><meta name="twitter:card" content="summary_large_image" /><meta name="twitter:site" content="@ODSC" /><meta name="twitter:label1" content="Est. reading time" /><meta name="twitter:data1" content="24 minutes" /> <script type="application/ld+json" class="yoast-schema-graph">{"@context":"https://schema.org","@graph":[{"@type":"WebPage","@id":"https://odsc.com/california/odsc-west-schedule-2023/","url":"https://odsc.com/california/odsc-west-schedule-2023/","name":"ODSC West Schedule 2024 | Open Data Science Conference","isPartOf":{"@id":"https://odsc.com/#website"},"datePublished":"2021-09-13T00:05:17+00:00","dateModified":"2024-10-31T22:07:23+00:00","description":"ODSC West 2022 West Preliminary Session Schedule","breadcrumb":{"@id":"https://odsc.com/california/odsc-west-schedule-2023/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https://odsc.com/california/odsc-west-schedule-2023/"]}]},{"@type":"BreadcrumbList","@id":"https://odsc.com/california/odsc-west-schedule-2023/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://odsc.com/"},{"@type":"ListItem","position":2,"name":"ODSC West 2024","item":"https://odsc.com/california/"},{"@type":"ListItem","position":3,"name":"ODSC West Schedule 2023"}]},{"@type":"WebSite","@id":"https://odsc.com/#website","url":"https://odsc.com/","name":"Open Data Science Conference","description":"Data Science Conferences","publisher":{"@id":"https://odsc.com/#organization"},"inLanguage":"en-US"},{"@type":"Organization","@id":"https://odsc.com/#organization","name":"ODSC","url":"https://odsc.com/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https://odsc.com/#/schema/logo/image/","url":"https://odsc.com/wp-content/uploads/2015/12/HIGHRESODSCLOGO.png","contentUrl":"https://odsc.com/wp-content/uploads/2015/12/HIGHRESODSCLOGO.png","width":991,"height":302,"caption":"ODSC"},"image":{"@id":"https://odsc.com/#/schema/logo/image/"},"sameAs":["https://www.facebook.com/OPENDATASCI","https://x.com/ODSC"]}]}</script> <!-- / Yoast SEO Premium plugin. --><link rel='dns-prefetch' href='//js.hs-scripts.com' /> <script type="text/javascript">var ajaxurl = 'https://odsc.com/wp-admin/admin-ajax.php';</script><link rel='stylesheet' id='bdp-widget-fontawesome-stylesheets-css' href='https://odsc.com/wp-content/plugins/blog-designer-pro/public/css/font-awesome.min.css' type='text/css' media='all' /> <style id='widget-for-eventbrite-api-display-eventbrite-events-style-inline-css' type='text/css'> </style> <style id='classic-theme-styles-inline-css' type='text/css'> /*! This file is auto-generated */ .wp-block-button__link{color:#fff;background-color:#32373c;border-radius:9999px;box-shadow:none;text-decoration:none;padding:calc(.667em + 2px) calc(1.333em + 2px);font-size:1.125em}.wp-block-file__button{background:#32373c;color:#fff;text-decoration:none} </style> <style id='global-styles-inline-css' type='text/css'> :root{--wp--preset--aspect-ratio--square: 1;--wp--preset--aspect-ratio--4-3: 4/3;--wp--preset--aspect-ratio--3-4: 3/4;--wp--preset--aspect-ratio--3-2: 3/2;--wp--preset--aspect-ratio--2-3: 2/3;--wp--preset--aspect-ratio--16-9: 16/9;--wp--preset--aspect-ratio--9-16: 9/16;--wp--preset--color--black: #000000;--wp--preset--color--cyan-bluish-gray: #abb8c3;--wp--preset--color--white: #ffffff;--wp--preset--color--pale-pink: #f78da7;--wp--preset--color--vivid-red: #cf2e2e;--wp--preset--color--luminous-vivid-orange: #ff6900;--wp--preset--color--luminous-vivid-amber: #fcb900;--wp--preset--color--light-green-cyan: #7bdcb5;--wp--preset--color--vivid-green-cyan: #00d084;--wp--preset--color--pale-cyan-blue: #8ed1fc;--wp--preset--color--vivid-cyan-blue: #0693e3;--wp--preset--color--vivid-purple: #9b51e0;--wp--preset--gradient--vivid-cyan-blue-to-vivid-purple: linear-gradient(135deg,rgba(6,147,227,1) 0%,rgb(155,81,224) 100%);--wp--preset--gradient--light-green-cyan-to-vivid-green-cyan: linear-gradient(135deg,rgb(122,220,180) 0%,rgb(0,208,130) 100%);--wp--preset--gradient--luminous-vivid-amber-to-luminous-vivid-orange: linear-gradient(135deg,rgba(252,185,0,1) 0%,rgba(255,105,0,1) 100%);--wp--preset--gradient--luminous-vivid-orange-to-vivid-red: linear-gradient(135deg,rgba(255,105,0,1) 0%,rgb(207,46,46) 100%);--wp--preset--gradient--very-light-gray-to-cyan-bluish-gray: linear-gradient(135deg,rgb(238,238,238) 0%,rgb(169,184,195) 100%);--wp--preset--gradient--cool-to-warm-spectrum: linear-gradient(135deg,rgb(74,234,220) 0%,rgb(151,120,209) 20%,rgb(207,42,186) 40%,rgb(238,44,130) 60%,rgb(251,105,98) 80%,rgb(254,248,76) 100%);--wp--preset--gradient--blush-light-purple: linear-gradient(135deg,rgb(255,206,236) 0%,rgb(152,150,240) 100%);--wp--preset--gradient--blush-bordeaux: linear-gradient(135deg,rgb(254,205,165) 0%,rgb(254,45,45) 50%,rgb(107,0,62) 100%);--wp--preset--gradient--luminous-dusk: linear-gradient(135deg,rgb(255,203,112) 0%,rgb(199,81,192) 50%,rgb(65,88,208) 100%);--wp--preset--gradient--pale-ocean: linear-gradient(135deg,rgb(255,245,203) 0%,rgb(182,227,212) 50%,rgb(51,167,181) 100%);--wp--preset--gradient--electric-grass: linear-gradient(135deg,rgb(202,248,128) 0%,rgb(113,206,126) 100%);--wp--preset--gradient--midnight: linear-gradient(135deg,rgb(2,3,129) 0%,rgb(40,116,252) 100%);--wp--preset--font-size--small: 13px;--wp--preset--font-size--medium: 20px;--wp--preset--font-size--large: 36px;--wp--preset--font-size--x-large: 42px;--wp--preset--spacing--20: 0.44rem;--wp--preset--spacing--30: 0.67rem;--wp--preset--spacing--40: 1rem;--wp--preset--spacing--50: 1.5rem;--wp--preset--spacing--60: 2.25rem;--wp--preset--spacing--70: 3.38rem;--wp--preset--spacing--80: 5.06rem;--wp--preset--shadow--natural: 6px 6px 9px rgba(0, 0, 0, 0.2);--wp--preset--shadow--deep: 12px 12px 50px rgba(0, 0, 0, 0.4);--wp--preset--shadow--sharp: 6px 6px 0px rgba(0, 0, 0, 0.2);--wp--preset--shadow--outlined: 6px 6px 0px -3px rgba(255, 255, 255, 1), 6px 6px rgba(0, 0, 0, 1);--wp--preset--shadow--crisp: 6px 6px 0px rgba(0, 0, 0, 1);}:where(.is-layout-flex){gap: 0.5em;}:where(.is-layout-grid){gap: 0.5em;}body .is-layout-flex{display: flex;}.is-layout-flex{flex-wrap: wrap;align-items: center;}.is-layout-flex > :is(*, div){margin: 0;}body .is-layout-grid{display: grid;}.is-layout-grid > :is(*, div){margin: 0;}:where(.wp-block-columns.is-layout-flex){gap: 2em;}:where(.wp-block-columns.is-layout-grid){gap: 2em;}:where(.wp-block-post-template.is-layout-flex){gap: 1.25em;}:where(.wp-block-post-template.is-layout-grid){gap: 1.25em;}.has-black-color{color: var(--wp--preset--color--black) !important;}.has-cyan-bluish-gray-color{color: var(--wp--preset--color--cyan-bluish-gray) !important;}.has-white-color{color: var(--wp--preset--color--white) !important;}.has-pale-pink-color{color: var(--wp--preset--color--pale-pink) !important;}.has-vivid-red-color{color: var(--wp--preset--color--vivid-red) !important;}.has-luminous-vivid-orange-color{color: var(--wp--preset--color--luminous-vivid-orange) !important;}.has-luminous-vivid-amber-color{color: var(--wp--preset--color--luminous-vivid-amber) !important;}.has-light-green-cyan-color{color: var(--wp--preset--color--light-green-cyan) !important;}.has-vivid-green-cyan-color{color: var(--wp--preset--color--vivid-green-cyan) !important;}.has-pale-cyan-blue-color{color: var(--wp--preset--color--pale-cyan-blue) !important;}.has-vivid-cyan-blue-color{color: var(--wp--preset--color--vivid-cyan-blue) !important;}.has-vivid-purple-color{color: var(--wp--preset--color--vivid-purple) !important;}.has-black-background-color{background-color: var(--wp--preset--color--black) !important;}.has-cyan-bluish-gray-background-color{background-color: var(--wp--preset--color--cyan-bluish-gray) !important;}.has-white-background-color{background-color: var(--wp--preset--color--white) !important;}.has-pale-pink-background-color{background-color: var(--wp--preset--color--pale-pink) !important;}.has-vivid-red-background-color{background-color: var(--wp--preset--color--vivid-red) !important;}.has-luminous-vivid-orange-background-color{background-color: var(--wp--preset--color--luminous-vivid-orange) !important;}.has-luminous-vivid-amber-background-color{background-color: var(--wp--preset--color--luminous-vivid-amber) !important;}.has-light-green-cyan-background-color{background-color: var(--wp--preset--color--light-green-cyan) !important;}.has-vivid-green-cyan-background-color{background-color: var(--wp--preset--color--vivid-green-cyan) !important;}.has-pale-cyan-blue-background-color{background-color: var(--wp--preset--color--pale-cyan-blue) !important;}.has-vivid-cyan-blue-background-color{background-color: var(--wp--preset--color--vivid-cyan-blue) !important;}.has-vivid-purple-background-color{background-color: var(--wp--preset--color--vivid-purple) !important;}.has-black-border-color{border-color: var(--wp--preset--color--black) !important;}.has-cyan-bluish-gray-border-color{border-color: var(--wp--preset--color--cyan-bluish-gray) !important;}.has-white-border-color{border-color: var(--wp--preset--color--white) !important;}.has-pale-pink-border-color{border-color: var(--wp--preset--color--pale-pink) !important;}.has-vivid-red-border-color{border-color: var(--wp--preset--color--vivid-red) !important;}.has-luminous-vivid-orange-border-color{border-color: var(--wp--preset--color--luminous-vivid-orange) !important;}.has-luminous-vivid-amber-border-color{border-color: var(--wp--preset--color--luminous-vivid-amber) !important;}.has-light-green-cyan-border-color{border-color: var(--wp--preset--color--light-green-cyan) !important;}.has-vivid-green-cyan-border-color{border-color: var(--wp--preset--color--vivid-green-cyan) !important;}.has-pale-cyan-blue-border-color{border-color: var(--wp--preset--color--pale-cyan-blue) !important;}.has-vivid-cyan-blue-border-color{border-color: var(--wp--preset--color--vivid-cyan-blue) !important;}.has-vivid-purple-border-color{border-color: var(--wp--preset--color--vivid-purple) !important;}.has-vivid-cyan-blue-to-vivid-purple-gradient-background{background: var(--wp--preset--gradient--vivid-cyan-blue-to-vivid-purple) !important;}.has-light-green-cyan-to-vivid-green-cyan-gradient-background{background: var(--wp--preset--gradient--light-green-cyan-to-vivid-green-cyan) !important;}.has-luminous-vivid-amber-to-luminous-vivid-orange-gradient-background{background: var(--wp--preset--gradient--luminous-vivid-amber-to-luminous-vivid-orange) !important;}.has-luminous-vivid-orange-to-vivid-red-gradient-background{background: var(--wp--preset--gradient--luminous-vivid-orange-to-vivid-red) !important;}.has-very-light-gray-to-cyan-bluish-gray-gradient-background{background: var(--wp--preset--gradient--very-light-gray-to-cyan-bluish-gray) !important;}.has-cool-to-warm-spectrum-gradient-background{background: var(--wp--preset--gradient--cool-to-warm-spectrum) !important;}.has-blush-light-purple-gradient-background{background: var(--wp--preset--gradient--blush-light-purple) !important;}.has-blush-bordeaux-gradient-background{background: var(--wp--preset--gradient--blush-bordeaux) !important;}.has-luminous-dusk-gradient-background{background: var(--wp--preset--gradient--luminous-dusk) !important;}.has-pale-ocean-gradient-background{background: var(--wp--preset--gradient--pale-ocean) !important;}.has-electric-grass-gradient-background{background: var(--wp--preset--gradient--electric-grass) !important;}.has-midnight-gradient-background{background: var(--wp--preset--gradient--midnight) !important;}.has-small-font-size{font-size: var(--wp--preset--font-size--small) !important;}.has-medium-font-size{font-size: var(--wp--preset--font-size--medium) !important;}.has-large-font-size{font-size: var(--wp--preset--font-size--large) !important;}.has-x-large-font-size{font-size: var(--wp--preset--font-size--x-large) !important;} :where(.wp-block-post-template.is-layout-flex){gap: 1.25em;}:where(.wp-block-post-template.is-layout-grid){gap: 1.25em;} :where(.wp-block-columns.is-layout-flex){gap: 2em;}:where(.wp-block-columns.is-layout-grid){gap: 2em;} :root :where(.wp-block-pullquote){font-size: 1.5em;line-height: 1.6;} </style> <style id='wcs-timetable-inline-css' type='text/css'> .wcs-single__action .wcs-btn--action{color:rgba( 255,255,255,1);background-color:#BD322C} </style> <style id='wpex-timeline-dark-css-inline-css' type='text/css'> .wpex-timeline > li .wpex-timeline-icon .fa{font-weight: normal;} </style> <style id='wpgb-head-inline-css' type='text/css'> .wp-grid-builder:not(.wpgb-template),.wpgb-facet{opacity:0.01}.wpgb-facet fieldset{margin:0;padding:0;border:none;outline:none;box-shadow:none}.wpgb-facet fieldset:last-child{margin-bottom:40px;}.wpgb-facet fieldset legend.wpgb-sr-only{height:1px;width:1px} </style> <!-- Google Analytics --> <script> (function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){ (i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o), m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m) })(window,document,'script','//www.google-analytics.com/analytics.js','ga'); ga('create', 'UA-69911961-1', 'auto'); ga('send', 'pageview'); </script> <!-- End Google Analytics --> <script type="text/javascript" src="https://odsc.com/wp-includes/js/jquery/jquery.min.js" id="jquery-core-js"></script> <script type="text/javascript" src="https://odsc.com/wp-includes/js/jquery/jquery-migrate.min.js" id="jquery-migrate-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-content/plugins/blog-designer-pro/public/js/imagesloaded.pkgd.min.js" id="bdp-imagesloaded-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-content/plugins/blog-designer-pro/public/js/isotope.pkgd.min.js" id="bdp_isotope_script-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-content/plugins/blog-designer-pro/public/js/ticker.min.js" id="bdp-ticker-js"></script> <script type="text/javascript" id="be-gdpr-js-extra"> /* <![CDATA[ */ var beGdprConcerns = {"youtube":{"label":"Youtube","description":"Consent to display content from YouTube.","required":false},"vimeo":{"label":"Vimeo","description":"Consent to display content from Vimeo.","required":false},"gmaps":{"label":"Google Maps","description":"Consent to display content from Google Maps.","required":false}}; /* ]]> */ </script> <script defer type="text/javascript" src="https://odsc.com/wp-content/uploads/siteground-optimizer-assets/be-gdpr.min.js" id="be-gdpr-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-content/plugins/svg-support/vendor/DOMPurify/DOMPurify.min.js" id="bodhi-dompurify-library-js"></script> <script type="text/javascript" id="widget-for-eventbrite-api-display-eventbrite-events-script-js-extra"> /* <![CDATA[ */ var wfea_freemius = {"current_plan":"PLAN_NAME","can_use_premium_code":"","is_plan_silver":"","is_plan_gold":"","is_plan_platinum":""}; var wfea_controls_meta = {"common":{"free":[{"name":"wfea_limit","label":"Number of Events to Display","type":"number","args":{"default":5}},{"name":"wfea_order_by","label":"Event Sort Order","type":"select2","args":{"options":{"asc":"Ascending Date","desc":"Descending Date","created_asc":"Created Date \u2013 ascending","created_desc":"Created Date \u2013 descending","published_asc":"EB Published Ascending","published_desc":"EB Published Descending"}}}]},"display":{"free":[{"name":"wfea_booknow","label":"Book Now button"},{"name":"wfea_date","label":"Event Date\/Time in Heading","args":{"condition":{"wfea_layout!":"short_date"}}},{"name":"wfea_excerpt","label":"Excerpt of Event Summary"},{"name":"wfea_thumb","label":"Image Display"},{"name":"wfea_thumb_original","label":"High Resolution Image","args":{"condition":{"wfea_thumb":["true"]}}},{"name":"wfea_readmore","label":"Read More Link on Excerpt","args":{"condition":{"wfea_layout!":["cal","cal_list"]}}}]},"enabling":{"free":[{"name":"wfea_newtab","label":"Link to EB in New Tab","args":{"default":"","condition":{"wfea_layout!":["cal","cal_list"]}}}]},"filtering":[],"selection":{"free":[{"name":"wfea_status","label":"Event Status","type":"select2","args":{"options":{"live":"Live","started":"Started","ended":"Ended","canceled":"Canceled","draft":"Draft","all":"All"},"default":"live","multiple":"true"}}]},"settings":{"free":[{"name":"wfea_booknow_text","label":"Book Now Wording","type":"text","args":{"default":"Register >>"}},{"name":"wfea_css_class","label":"Custom CSS Class","type":"text"},{"name":"wfea_cssID","label":"Custom CSS ID","type":"text"},{"name":"wfea_thumb_align","label":"Image Alignment","type":"select2","args":{"options":{"eaw-alignleft":"Left","eaw-alignright":"Right","eaw-aligncenter":"Center"},"condition":{"wfea_layout":"widget"},"default":"eaw-alignright"}},{"name":"wfea_length","label":"Length of Description","type":"number"},{"name":"wfea_thumb_width","label":"Image Width","type":"number","args":{"condition":{"wfea_layout":"widget"},"default":350,"max":""}},{"name":"wfea_thumb_default","label":"Default Image","type":"text"},{"name":"wfea_readmore_text","label":"Read More Wording","type":"text","args":{"default":"Read More >>","condition":{"wfea_layout!":["cal","cal_list"]}}}]},"ajaxurl":"https:\/\/odsc.com\/wp-admin\/admin-ajax.php","nonce":"4ff34ac970"}; /* ]]> */ </script> <script defer type="text/javascript" id="cookie-notice-front-js-before"> /* <![CDATA[ */ var cnArgs = {"ajaxUrl":"https:\/\/odsc.com\/wp-admin\/admin-ajax.php","nonce":"1ff2934577","hideEffect":"fade","position":"bottom","onScroll":true,"onScrollOffset":100,"onClick":false,"cookieName":"cookie_notice_accepted","cookieTime":2592000,"cookieTimeRejected":2592000,"globalCookie":false,"redirection":false,"cache":false,"revokeCookies":true,"revokeCookiesOpt":"manual"}; /* ]]> */ </script> <script defer type="text/javascript" src="https://odsc.com/wp-content/plugins/cookie-notice/js/front.min.js" id="cookie-notice-front-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-includes/js/jquery/ui/core.min.js" id="jquery-ui-core-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-content/uploads/siteground-optimizer-assets/bodhi_svg_inline.min.js" id="bodhi_svg_inline-js"></script> <script defer type="text/javascript" id="bodhi_svg_inline-js-after"> /* <![CDATA[ */ cssTarget={"Bodhi":"img.style-svg","ForceInlineSVG":"style-svg"};ForceInlineSVGActive="false";frontSanitizationEnabled="on"; /* ]]> */ </script> <script defer type="text/javascript" src="https://odsc.com/wp-content/plugins/tatsu/includes/typehub/public/js/webfont.min.js" id="webfontloader-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-content/uploads/siteground-optimizer-assets/oba_youtubepopup_plugin.min.js" id="oba_youtubepopup_plugin-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-content/uploads/siteground-optimizer-assets/oba_youtubepopup_activate.min.js" id="oba_youtubepopup_activate-js"></script> <script type="text/javascript" id="etn-public-pro-js-extra"> /* <![CDATA[ */ var etn_pro_public_object = {"ajax_url":"https:\/\/odsc.com\/wp-admin\/admin-ajax.php","location_map_nonce":"defad18d7a","scanner_nonce":"d54224dade","attendee_page_link":"https:\/\/odsc.com\/wp-admin\/edit.php?post_type=etn-attendee","scanner_common_msg":"Something went wrong! Please try again.","attendee_verification_style":"on","location_icon":"https:\/\/odsc.com\/wp-content\/plugins\/eventin-pro\/assets\/images\/location-icon.png","attendee_registration_option":"","event_expired_message":"This event has been expired."}; /* ]]> */ </script> <script defer type="text/javascript" src="https://odsc.com/wp-content/uploads/siteground-optimizer-assets/etn-public-pro.min.js" id="etn-public-pro-js"></script> <script type="text/javascript" id="etn-rsvp-public-js-extra"> /* <![CDATA[ */ var localized_rsvp_data = {"ajax_url":"https:\/\/odsc.com\/wp-admin\/admin-ajax.php","attendee_title":"Attendee"}; /* ]]> */ </script> <script defer type="text/javascript" src="https://odsc.com/wp-content/uploads/siteground-optimizer-assets/etn-rsvp-public.min.js" id="etn-rsvp-public-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-content/uploads/siteground-optimizer-assets/etn-google-public.min.js" id="etn-google-public-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-includes/js/jquery/ui/mouse.min.js" id="jquery-ui-mouse-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-includes/js/jquery/ui/resizable.min.js" id="jquery-ui-resizable-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-includes/js/jquery/ui/draggable.min.js" id="jquery-ui-draggable-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-includes/js/jquery/ui/controlgroup.min.js" id="jquery-ui-controlgroup-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-includes/js/jquery/ui/checkboxradio.min.js" id="jquery-ui-checkboxradio-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-includes/js/jquery/ui/button.min.js" id="jquery-ui-button-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-includes/js/jquery/ui/dialog.min.js" id="jquery-ui-dialog-js"></script> <script type="text/javascript" id="widget-for-eventbrite-api-js-extra"> /* <![CDATA[ */ var wfea_frontend = {"ajaxurl":"https:\/\/odsc.com\/wp-admin\/admin-ajax.php"}; /* ]]> */ </script> <script defer type="text/javascript" src="https://odsc.com/wp-content/uploads/siteground-optimizer-assets/widget-for-eventbrite-api.min.js" id="widget-for-eventbrite-api-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-content/themes/oshin/oshin/js/vendor/modernizr.min.js" id="modernizr-js"></script><link rel="https://api.w.org/" href="https://odsc.com/wp-json/" /><link rel="alternate" title="JSON" type="application/json" href="https://odsc.com/wp-json/wp/v2/pages/67937" /><link rel="EditURI" type="application/rsd+xml" title="RSD" href="https://odsc.com/xmlrpc.php?rsd" /><meta name="generator" content="WordPress 6.7" /><link rel='shortlink' href='https://odsc.com/?p=67937' /><link rel="alternate" title="oEmbed (JSON)" type="application/json+oembed" href="https://odsc.com/wp-json/oembed/1.0/embed?url=https%3A%2F%2Fodsc.com%2Fcalifornia%2Fodsc-west-schedule-2024%2F" /><link rel="alternate" title="oEmbed (XML)" type="text/xml+oembed" href="https://odsc.com/wp-json/oembed/1.0/embed?url=https%3A%2F%2Fodsc.com%2Fcalifornia%2Fodsc-west-schedule-2024%2F&format=xml" /><meta name="framework" content="Redux 4.1.29" /><!-- start Simple Custom CSS and JS --> <style type="text/css"> /* Add your CSS code here. For example: .example { color: red; } For brushing up on your CSS knowledge, check out http://www.w3schools.com/css/css_syntax.asp End of comment */ /*.single-portfolio .butonici {display:none!important}*/</style> <!-- end Simple Custom CSS and JS --> <!-- start Simple Custom CSS and JS --> <style type="text/css"> /* Add your CSS code here. For example: .example { color: red; } For brushing up on your CSS knowledge, check out http://www.w3schools.com/css/css_syntax.asp End of comment */ /*.group-content__price__table.c_5 .price__table__column.column--ticket:nth-child(4) .column__title{ background: #00bcdd!important; } .group-content__price__table.c_5 .price__table__column.column--ticket:nth-child(4) .column__title:after { border-top: 8px solid #00bcdd!important; } .group-content__price__table.c_5 .price__table__column.column--ticket:nth-child(4) .button { border-radius: 5px!important; background-color: #00bcdd!important; padding: 8px 32px!important; } .group-content__price__table.c_5 .price__table__column.column--ticket:nth-child(4) .column__option .fa-check { color:#00bcdd; } .group-content__price__table.c_5 .price__table__column.column--ticket:nth-child(4) .button:hover { color:#15244a!important; } */ .secfocus { white-space: nowrap; } .etn-content-item{color:#e6d7d7!important} .schedule-tab-2 .etn-schedule-speaker-title { color: white!important; } .schedule-tab-2 .etn-schedule-single-speaker span { display: block; color: #1a9fe5; } .etn-icon.fa-linkedin:before { content: "\e904"!important; } .single-etn-speaker .etn-etn-icon.fa-linkedin:before{display:none!important} .single-etn-speaker .etn-schedule-location {display:none!important} .etn-event-item .etn-event-location, .etn-event-tag-list a:hover, .etn-schedule-wrap .etn-schedule-info .etn-schedule-time { color: #0057c1; /*display: none!important;*/ } .firstfocus{background:blue;padding: 7px 15px;margin: 5px;margin-right: 3px;border-radius: 10px;} .secfocus{background: #00bcdd;padding: 7px 15px;margin-left: 3px;border-radius: 10px;} .thirdtfocus{background: #00bcdd;padding: 7px 15px;margin-left: 3px;border-radius: 10px;} .locationvirtlive{margin-top: 20px;margin-left: 6px;font-weight: bold;font-size: 18px;} .etn-schedule-wrap .etn-schedule-info .etn-schedule-time { margin-bottom: 0px!important; } .etn-schedule-wrap .etn-schedule-info .etn-schedule-time { background: #1a9fe5; padding: 5px 18px!important;} .etn-schedule-wrap .etn-single-schedule-item { border-bottom: 1px solid #eaeaea; padding: 20px 0!important; margin: 0; } .etn-schedule-wrap .etn-schedule-info .etn-schedule-location { display: block; padding-left: 0px; font-size: .875rem; line-height: 2.25rem!important; position: relative; padding-top: 10px; } .etn-schedule-wrap .etn-schedule-info .etn-schedule-location i{display:none!important;} .etn-schedule-wrap .etn-schedule-info .etn-schedule-location { display: block; padding-left: 0px; font-size: .875rem; line-height: 1.25rem; position: relative; padding-top: 10px; } .etn-event-item .etn-event-location, .etn-event-tag-list a:hover, .etn-schedule-wrap .etn-schedule-info .etn-schedule-time { color: #0057c1; /* display: none;*/ } .etn-location:before { content: "\e916"!important; } .schedule-tab-wrapper .etn-nav { padding: 0; margin: 0 0 40px; margin: 0 auto; display: table; padding: 30px 0px; /* display: none!important;*/ } .schedule-tab-wrapper .etn-schedule-speaker .etn-schedule-single-speaker .etn-schedule-speaker-title { position: relative; left: 0; top: 0; padding: 0; background: #5d78ff; color: #fff; font-size: 20px; line-height: 30px; padding: 3px 10px; transition: all .4s ease; -webkit-transition: all .4s ease; -moz-transition: all .4s ease; -ms-transition: all .4s ease; opacity: 1; visibility: visible; margin: 0 auto; white-space: nowrap; display: table; } .schedule-tab-wrapper .etn-schedule-speaker .etn-schedule-single-speaker img { width: 200px; height: 200px; object-fit: cover; } .etn-schedule-wrap .etn-schedule-info .etn-schedule-time { background: #1a9fe5; padding: 10px 18px; border-radius: 36px; -webkit-border-radius: 36px; -ms-border-radius: 36px; color: white!important; font-size: 15px; font-weight: 700; display: inline-block; text-transform: uppercase; margin-bottom: 10px; } .schedule-tab-2 .etn-accordion-heading i { padding: 8px 10px; background: #ffeff5; color: #189fe5; border-radius: 5px; -webkit-border-radius: 5px; -ms-border-radius: 5px; } .etn-acccordion-contents {padding-right: 55px;} .schedule-tab-2 .etn-schedule-single-speaker img { border-radius: 5px; -webkit-border-radius: 5px; -ms-border-radius: 5px; width: 100px; height: 100px; object-fit: cover; } .schedule-tab-2 .etn-schedule-speaker-designation { font-size: 17px; line-height: 24px; } .schedule-tab-2 .etn-schedule-speaker-title { font-size: 18px; font-weight: 500; line-height: 28px; color: white; } .schedule-tab-2 .etn-schedule-single-speaker { margin-right: 20px; display: inline-block; text-align: left; } .etn-schedule-info{color: white;} .etn-schedule-wrap .etn-schedule-content .etn-title { margin-bottom: 10px; color: white; } .schedule-tab-wrapper .etn-nav li a .etn-day { font-size: .9375rem; color: #ffffff; font-weight: 500; text-align: center; } .schedule-tab-wrapper .etn-nav li a { font-size: 1.25rem; font-weight: 700; color: #ffffff;} .schedule-tab-wrapper .etn-nav { padding: 0; margin: 0 0 40px; margin: 0 auto; display: table;padding: 30px 0px; }</style> <!-- end Simple Custom CSS and JS --> <!-- start Simple Custom CSS and JS --> <style type="text/css"> /* Add your CSS code here. For example: .example { color: red; } For brushing up on your CSS knowledge, check out http://www.w3schools.com/css/css_syntax.asp End of comment */ .wp-grid-builder .wpgb-sidebar { flex-basis: 300px; min-width: 0; padding-left: 40px; } .wp-grid-builder .wpgb-scheme-dark .wpgb-idle-scheme-2, .wp-grid-builder .wpgb-scheme-dark [class^="wpgb-block-"].wpgb-hover-scheme-2:hover { color: #565656; margin-bottom: 30px; } .tcode-event-schedule .scheduled-days .scheduled-day .row-day { font-size: 18px!important; padding-top: 26px; line-break: normal; } @media only screen and (min-width: 767px) { .group-content__price__table.c_6 .price__table__column.column--ticket { width: 14.6%!important; } .group-content__price__table.c_6 .price__table__column.column--options { border-color: transparent; width: 280px!important; position: absolute; margin-left: -156px!important; } .group-content__price__table.c_6 { display: table; width: 100%!important; max-width: 100%!important; margin: 0 auto; margin-top: 35px; position: relative; padding-left: 150px; } .group-content__price__table.c_5 .price__table__column.column--options { border-color: transparent; width: 279px!important; position: absolute; margin-left: -149px!important; } } </style> <!-- end Simple Custom CSS and JS --> <!-- start Simple Custom CSS and JS --> <style type="text/css"> /* Add your CSS code here. For example: .example { color: red; } For brushing up on your CSS knowledge, check out http://www.w3schools.com/css/css_syntax.asp End of comment */ .envira-tags-filter-active{ border-radius: 0!important; background-color: transparent!important; color: black!important!important; border: 4px solid black!important; PADDING: 10PX 20PX!important;} .envira-tags-filter-link{ border-radius: 0!important; background-color: transparent!important; color: black!important; } .envira-tags-filter-list{ display: table!important; margin: 0 auto!important; padding-bottom: 40px!important;} .filters.rounded .sort { padding: 3px 20px; color: black!important; font-size: 18px; } .filters.rounded .current_choice { border-radius: 0!important; background-color: transparent!important; color: black!important; border: 4px solid black!important; } #portdark .thumb-wrap{pointer-events:none!important; cursor:normal!important}</style> <!-- end Simple Custom CSS and JS --> <!-- start Simple Custom CSS and JS --> <style type="text/css"> /* Add your CSS code here. For example: .example { color: red; } For brushing up on your CSS knowledge, check out http://www.w3schools.com/css/css_syntax.asp End of comment */ #tree-container { display: flex; justify-content: center; } .node { cursor: pointer; } .overlay { background-color: black!important; } .node circle { stroke: steelblue; stroke-width: 1.5px; } .node text { font-size: 14px; font-family: sans-serif; } .link { fill: none; } div.tooltip { position: absolute; padding: 10px 15px; font: 12px sans-serif; background: lightsteelblue; border: 0px; border-radius: 8px; z-index: 1; box-shadow: 0 5px 10px rgba(0,0,0,0.2); } div.tooltip:empty { padding: 0; } div.tooltip p { margin: 5px 0 0; } </style> <!-- end Simple Custom CSS and JS --> <!-- start Simple Custom CSS and JS --> <style type="text/css"> /* Add your CSS code here. For example: .example { color: red; } For brushing up on your CSS knowledge, check out http://www.w3schools.com/css/css_syntax.asp End of comment */ #ptsTableInitEditHtmlDlg{width:100%!important} .woocommerce div.product div.images, .woocommerce #content div.product div.images, .woocommerce-page div.product div.images, .woocommerce-page #content div.product div.images { margin-top: 0px; } .woocommerce div.product div.summary, .woocommerce #content div.product div.summary, .woocommerce-page div.product div.summary, .woocommerce-page #content div.product div.summary { margin-bottom: 2em; width: 48%; float: right; margin-top: 50px; } .woocommerce .product-single-boxed-content { margin-bottom: 0px; } .woocommerce .product-single-boxed-content { margin-bottom: 0px; background: #ebeff6; padding-top: 0; margin-bottom: 40px; } .woocommerce div.product form.cart .button, .woocommerce #content div.product form.cart .button, .woocommerce-page div.product form.cart .button, .woocommerce-page #content div.product form.cart .button { vertical-align: middle; float: left; background: #00bcdd!important; border: #00bcdd!important; color: white!important; } .woocommerce div.product form.cart, .woocommerce #content div.product form.cart, .woocommerce-page div.product form.cart, .woocommerce-page #content div.product form.cart { margin-bottom: 35px; margin-top: 20px; text-transform: uppercase; } .woocommerce-tabs{display:none!important} .woocommerce-variation-price .price{display:block!important} .entry-summary .price, .product_meta{display:none} .woocommerce-page.single.single-product #content div.product h1.product_title.entry-title{ font: 700 35px "Source Sans Pro","Open Sans","Arial",sans-serif!important; color: #515151!important; line-height: 42px!important; letter-spacing: 1px!important; text-transform: none!important; } .product-single-boxed-content .custom-share-button{display:none!important}</style> <!-- end Simple Custom CSS and JS --> <!-- start Simple Custom CSS and JS --> <style type="text/css"> /* Add your CSS code here. For example: .example { color: red; } For brushing up on your CSS knowledge, check out http://www.w3schools.com/css/css_syntax.asp End of comment */ /* Add your CSS code here. For example: .example { color: red; } For brushing up on your CSS knowledge, check out http://www.w3schools.com/css/css_syntax.asp End of comment */ .widgets_on_page>ul, .widgets_on_page>ul li { list-style: none; background: 0 0; padding-left: 0px; } .hbspt-form input[type="email"] {width: 100%!important;} .poster #navigation-right-side { width: 55%!important; padding-left:0px!important; } .poster #header-wrap .logo{display:none!important} .poster #header-top-bar{display:none!important} .poster #navigation-left-side a{display:none!important} .tatsu-divider-wrap.tatsu-animate{ visibility: visible!important; } .tatsu-shortcode-module .svg_container.big { padding: 20px; width: 80px; height: 80px; } a.glink img { margin: 0px 3px; } .menu-item-gtranslate:after{display:none!important} .paded{margin-top:40px!important; margin-bottom:40px!important;} #popmake-47777 .vc_column-inner{padding-top:0px!important;} #popmake-47777 .wpb_single_image.vc_align_left{margin-bottom:0px!important} #popmake-47777{ /* padding: 18px; border-radius: 0px!important; border: none!important; box-shadow: 0px 0px 30px 0px rgba( 2, 2, 2, 1.00 )!important; background-image: url(https://staging6.odsc.com/wp-content/uploads/2020/02/popup-2.jpg)!important;*/ padding-left: 0px!important; padding-top: 0px!important; padding-bottom: 0px!important;} .popi2 input[type="submit"] { background: #0072bc!important; color: white!important; text-transform: uppercase!important; font-size: 27px!important; font-weight: normal!important; padding: 19px 25px; text-align: center; width: 74%; margin: 0 auto; display: table; } .popi input[type="submit"] {background: #f7941d!important; color: white!important; text-transform: uppercase!important; font-size: 27px!important; font-weight: normal!important; padding: 19px 25px;} .hs-form-required{display:none!important} .texs{ border: 3px solid white; padding: 50px 40px;} /*.popmake-content p:first-child{display:none!important}*/ #popmake-47580{ padding: 18px; border-radius: 0px!important; border: none!important; box-shadow: 0px 0px 30px 0px rgba( 2, 2, 2, 1.00 )!important; /* background-color: rgba( 255, 255, 255, 1.00 ); */ background-image: url(https://staging6.odsc.com/wp-content/uploads/2020/02/popup-4.jpg)!important; } .icon-icon_plus:before { content: "\e806"; color: white; } #header .be-mobile-menu-icon span { background-color: #ffffff!important; } @media only screen and (max-width: 476px) { .schedule-tab-wrapper .etn-nav{display:none!important;} .ui-tabs .ui-tabs-nav li a { border-bottom: 2px solid; color: #1a9fe5; font-size: 30px; font-size: 20px!important; line-height: 30px!important; } .schedule-tab-2 .etn-schedule-single-speaker img { border-radius: 5px; -webkit-border-radius: 5px; -ms-border-radius: 5px; width: 50px; height: 50px; object-fit: cover; } } @media only screen and (max-width: 767px) { .etn-acccordion-contents { padding-right: 0px!important; } .nomob{display:none!important} #icon_wrapper { position: fixed; display:none!important; right: 0px; z-index: 99999; } .herojski{left: 0px!important;} } .column__option .fa-check{ color: #0054a6;} .butonici a{width:100%!important; padding-left:unset; padding-right:unset;} h6.gallery-side-heading { font-size: 26px!important; line-height: 32px; } .group-content__price__table { display: table; width: 100%; max-width: 920px; margin: 0 auto; margin-top: 35px; position: relative; padding-left: 150px; } .single-portfolio .onlyport{display:block!important;width: 100%;} .portfolio-details{padding: 0px 27px;} a.custom-share-button{font-size:20px!important} h6.gallery-side-heading { font-size: 26px!important; } .speaker-social-icons i{ font-size: 49px!important; width: 45px; height: 45px; line-height: 45px; margin: 0 auto; display: table; } .speaker-details-circle img { margin: 0 auto; display: table; } .speaks .tatsu-column{margin-bottom:10px!important;} .speaks p{margin-bottom: 20px!important;} .speaks .flip-wrap{width: 100px; float: right;} .speaks .thumb-overlay{display:none!important} .speaks .thumb-bg { display: none; } .single-post .post-title, .single-post .post-date-wrap{margin-top:40px;} .awsm-grid-wrapper .awsm-grid>.awsm-grid-card { min-height: 425px; } .awsm-personal-info h3{margin-bottom: 10px!important;} figcaption .awsm-personal-info{ border-bottom:none!important;} .awsm-personal-info{ padding-top: 20px; border-bottom: 1px solid black; padding-bottom: 15px;} .awsm-personal-info .comptit, .jobpos{ text-transform: uppercase; font-size: .9em; line-height: 1.9; display: block; color: black; } .jobpos{ font-style: italic;} #page-content h3{ font-family: Roboto; font-weight: 300;} p span, .tatsu-button-wrap a{ font-family: Roboto; font-weight: 300;} h5 span, h4{font-family: Roboto!important;} h2 span, .tatsu-text-inner, p, span, strong, h1, h2, h3, h4, h5 {font-family: Roboto!important;} h3{font-weight:700!important} #page-content li{font-family: Roboto!important; } .single-post .hero-section-wrap{height:400px;} .style1-blog .article-details, .style5-blog .article-details, .style6-blog .article-details { padding: 0px 0 0!important; } .single-post .content{padding-top:40px!important;} .single-post .tagcloud{display:none!important;} .post-template-default #header-wrap{display:block!important;} .post-title a{ font: 700 55px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #515151; line-height: 70px; letter-spacing: 0px; text-transform: none;} .style2-blog .post-nav, .style3-blog .post-nav, .style5-blog .post-nav, .style6-blog .post-nav{ display:none!important; } .single-post .section-overlay{ opacity: 0.5!important;} .single-post select{ width: 99%!important;} .single-post .hbspt-form input[type="text"]{ width: 95%!important;} .single-post .stacked .actions { margin-left: 0!important;margin-top:0px!important;padding-top:0px!important} .single-post .hs-button { display: table!important; margin: 0 auto!important; } .single-post .hbspt-form input[type="text"] { padding-bottom: 0px; width: 95%!important; } .single-post .hs-field-desc { width: 100%; color: #aaa; margin: 0px 0px 5px 150px; font-size: 11px; font-family: "Helvetica Neue",Helvetica,Arial,sans-serif; line-height: 15px; text-align: center; } .single-post .hs-form-required{display:none!important;} .single-post .hbspt-form { border: 1px solid #1fbdef!important; padding: 10px; } .single-post .hs-input {margin-bottom:0px!important} .single-post .hs-richtext h5 { font-family: Open Sans, Arial, sans-serif; font-size: 20px; text-align: center; color: #1fbdef; } #navigation-left-side { padding-right: 90px; } #navigation-right-side { padding-left: 90px; } #navigation-left-side a, #navigation-right-side a, #navigation a { padding: 0 10px!important;} .ui-accordion .ui-accordion-header.ui-state-active { background: #2CB5E9!important; } .ui-accordion .ui-accordion-header.ui-state-default { background: #428BCA; } body .ui-accordion .ui-accordion-header { margin-top: 1px; } body .ui-accordion .ui-accordion-header { padding: 14px 18px; color: #fff; font-size: 18px; } body .ui-accordion-header-icon.ui-icon.ui-icon-triangle-1-s, body .ui-accordion .ui-accordion-header.ui-state-active:hover .ui-accordion-header-icon.ui-icon.ui-icon-triangle-1-s { width: 24px; height: 24px; float: right!important; background: url(https://staging6.odsc.com/wp-content/themes/oshin/images/rounded_arrow_up.png) no-repeat center center transparent; } body .ui-accordion .ui-accordion-header .ui-accordion-header-icon { position: relative; left: auto; right: auto; top: 0; margin-top: 0; } .ui-icon.ui-icon-triangle-1-e, .ui-accordion .ui-accordion-header.ui-state-default:hover .ui-icon.ui-icon-triangle-1-e { width: 24px; height: 24px; float: right; background: url(https://staging6.odsc.com/wp-content/themes/oshin/images/rounded_arrow.png) no-repeat center center transparent; } .svg_container.big.handshake img { max-width: none; width: 56px; } p.date { color: #15c2e0; font-weight: bold; text-transform: uppercase; margin-bottom: 0; } .svg_container.big.handshake { padding: 25px 30px 20px 15px; } .svg_container.blue, .blue .svg_container { background: rgba(0, 188, 221, 0.32); } .svg_container.big { padding: 20px; width: 40px; height: 40px; } .svg_container { width: 28px; height: 28px; float: left; padding: 14px; margin-top: 0; margin-right: 14px; border-radius: 50%; }</style> <!-- end Simple Custom CSS and JS --> <!-- start Simple Custom CSS and JS --> <style type="text/css"> /* Add your CSS code here. For example: .example { color: red; } For brushing up on your CSS knowledge, check out http://www.w3schools.com/css/css_syntax.asp End of comment */ #ptsTableInitEditHtmlDlg{width:100%!important} #odsc-tickets .groups__group { } @media only screen and (max-width: 960px){ body.sticky-header #header #header-inner-wrap.no-transparent { position: fixed!important; left: 0!important; right: 0 !important; top: 0px !important; }} .kolumnspec select { width: 100%!important; font: 400 18px "Source Sans Pro","Open Sans","Arial",sans-serif!important; border:none!important; height:41px!important; } .newslee input[type="email"]{ width: 100%!important;} .kolumnspec ul { list-style: none!important; } #odsc-tickets .groups__group:hover { background: #f7f5f5; } /* Timetable */ @media only screen and (min-width: 992px){ .tcode-event-schedule .scheduled-event .artist-image { height: 92px; width: 92px; } } .artist-image .img-responsive{border-radius: 50%;} @media only screen and (max-width: 767px){ .klasicc .tatsu-eq-cols .tatsu-column { min-height: 270px!important; } } /*register page*/ @media only screen and (max-width: 460px) { .fixe span{line-height:55px;} } .mobmenu{display:none!important;} .mobmenu-push-wrap, body.mob-menu-slideout-over { padding-top: 0px!important; } @media only screen and (max-width: 964px) { .page-id-161 #header-inner-wrap{ /* display:none!important;*/ } .page-id-161 .mobmenu{display:block!important;} .page-id-161 .mobmenu-push-wrap, .page-id-161 body.mob-menu-slideout-over { padding-top: 40px!important; } } .mptt-shortcode-wrapper .mptt-shortcode-table{ color: white; } .mptt-shortcode-wrapper .mptt-shortcode-table tr.mptt-shortcode-row th{ background-color: transparent; color: white; padding: 1rem; } .mptt-shortcode-wrapper .mptt-shortcode-table tbody tr:nth-child(2n+2) { background-color: transparent; } .mptt-shortcode-wrapper .mptt-shortcode-table.mptt-theme-mode tbody td.event{ background-color: transparent; } .mptt-shortcode-wrapper .mptt-shortcode-table tbody div[data-color="#48AFDE"] .event-subtitle, .mptt-shortcode-wrapper .mptt-shortcode-table tbody div[data-color="#48AFDE"] .timeslot{ color: white !important; } .mptt-shortcode-wrapper .mptt-shortcode-table tbody .mptt-shortcode-event.mptt-event-vertical-top[colspan="4"] .mptt-event-container{ background-color: transparent !important; justify-content: center; } .mptt-shortcode-wrapper .mptt-shortcode-table tbody .mptt-event-container .event-title{ font-weight: bold; } .mptt-shortcode-wrapper .mptt-shortcode-table tbody [colspan="4"] .event-title{ font-size: 24px; } .mptt-shortcode-wrapper .mptt-shortcode-table tbody .mptt-event-container:hover .event-title{ text-decoration: none !important; } .mptt-shortcode-wrapper .mptt-shortcode-event.mptt-event-vertical-top[data-column-id="17418"] .mptt-event-container, .mptt-shortcode-wrapper .mptt-shortcode-event.mptt-event-vertical-top[data-column-id="17423"] .mptt-event-container{ background-color: rgba(105, 217, 226, 0.15); } .mptt-shortcode-wrapper .mptt-shortcode-event.mptt-event-vertical-top[data-column-id="17418"] .mptt-event-container:hover, .mptt-shortcode-wrapper .mptt-shortcode-event.mptt-event-vertical-top[data-column-id="17423"] .mptt-event-container:hover{ background-color: rgba(105, 217, 226, .25); } .mptt-shortcode-wrapper .mptt-shortcode-event.mptt-event-vertical-top[data-column-id="17424"] .mptt-event-container, .mptt-shortcode-wrapper .mptt-shortcode-event.mptt-event-vertical-top[data-column-id="17425"] .mptt-event-container{ background-color: rgba(226, 152, 178, 0.15); } .mptt-shortcode-wrapper .mptt-shortcode-event.mptt-event-vertical-top[data-column-id="17424"] .mptt-event-container:hover, .mptt-shortcode-wrapper .mptt-shortcode-event.mptt-event-vertical-top[data-column-id="17425"] .mptt-event-container:hover{ background-color: rgba(226, 152, 178, .25); } .mptt-shortcode-list { color: white; } [data-event-id="17452"] .timeslot{ display: none; } .mptt-shortcode-wrapper .mptt-shortcode-table tbody .mptt-event-container[data-event-id="17452"] .event-subtitle { font-size: .9em; } .mfp-content { position: relative; background-color: #FFF; padding: 40px 20px; width: 90%; max-width: 700px; margin: 20px auto; } .mfp-content h4{ text-align: center; } .mfp-content .timeslots-title,.mfp-content .timeslot{ display: none; } .mfp-preloader{ color: black; } .mfp-content{ position: relative; background-color: #FFF!important; padding: 40px 20px; width: 90%; max-width: 700px; margin: 20px auto;} .mptt-event-container.id-56 a{ font-weight:bold; color: rgb(0, 188, 221)!important; } .mptt-event-container.id-56 a::before{ } .mptt-event-container.id-63,.mptt-event-container.id-64 { background-image: url(https://staging6.odsc.com/wp-content/uploads/2018/05/DinnerDS.jpg);} .mptt-event-container.id-121,.mptt-event-container.id-64 { background-image: url(https://staging6.odsc.com/wp-content/uploads/2017/09/Data-Robot-Networking-Party1.jpg);} .linkic{width: 100%; height: 100%; position: absolute;} .overl{ position: absolute; top: 35%; background: url(https://staging6.odsc.com/wp-content/uploads/2018/05/play-white.svg); width: 20%; background-repeat: no-repeat; left: 40%; height: 100%;} #hu-header-mobile .hu-header-logo { max-height: 75px; margin: 41px auto 0; display: none!important; } .agendica tbody tr:nth-child(2n+2) { background-color: transparent!important; } .agendica tr.mptt-shortcode-row th { background-color: transparent!important; } .agendica tbody td.event { background-color: transparent!important; } .agendica .mptt-shortcode-hours{color:white!important} .agendica th{color:white!important} .agendica .timeslot{color:white!important} .agendica .event-subtitle{color:white!important} .agendica .event-title{color:white!important} .price__table__column .column__header { height: 250px!important; } .price__table__column .column__header .column__desc { font-size: 15px!important; line-height: 20px!important; } .faq_categories .attending { background: url(images/attendee.svg) no-repeat center 15px #fff; background-size: 37px; } .faq_categories .attending.activated { background: url(images/attendee_light.svg) no-repeat center 15px #fff; background-size: 37px; } .faq_categories .hiring { background: url(images/hiring.svg) no-repeat center 15px #fff; background-size: 37px; } .faq_categories .hiring.activated { background: url(images/hiring_light.svg) no-repeat center 15px #fff; background-size: 37px; } .faq_categories .videos { background: url(images/videos.svg) no-repeat center 15px #fff; background-size: 37px; } .faq_categories .videos.activated { background: url(images/videos_light.svg) no-repeat center 15px #fff; background-size: 37px; } .faq_categories .speakers { background: url(images/speaker.svg) no-repeat center 15px #fff; background-size: 37px; } .faq_categories .speakers.activated { background: url(images/speaker_light.svg) no-repeat center 15px #fff; background-size: 37px; } .faq_categories .volunteers { background: url(images/volunteer.svg) no-repeat center 15px #fff; background-size: 37px; } .faq_categories .volunteers.activated { background: url(images/volunteer_light.svg) no-repeat center 15px #fff; background-size: 37px; } .faq_categories .sponsor { background: url(images/sponsor.svg) no-repeat center 15px #fff; background-size: 37px; } .faq_categories .sponsor.activated { background: url(images/sponsor_light.svg) no-repeat center 15px #fff; background-size: 37px; } .faq_categories .group-tickets { background: url(images/group.svg) no-repeat center 12px #fff; background-size: 27px; } .faq_categories .ticket-type { background: url(images/ticket.svg) no-repeat center 12px #fff; background-size: 27px; } .faq_categories .schedule, .faq_categories .odsc { background: url(images/schedule.svg) no-repeat center 12px #fff; background-size: 27px; } .faq_categories .refund-policy { background: url(images/refund.svg) no-repeat center 12px #fff; background-size: 27px; } .faq_categories .trainings { background: url(images/training.svg) no-repeat center 12px #fff; background-size: 27px; } .faq_categories .app-info { background: url(images/mobile.svg) no-repeat center 12px #fff; background-size: 27px; } .faq_categories .scholarships { background: url(images/scholarship.svg) no-repeat center 12px #fff; background-size: 27px; } .faq_categories .promotions-pricing { background: url(images/pricing.svg) no-repeat center 12px #fff; background-size: 27px; } .faq_categories .odsc-europe-2017, .faq_categories .odsc-west-2017, .faq_categories .odsc-east-2017 { background: url(images/logo.png) no-repeat center 10px #fff; background-size: 37px; padding-top: 35px; } .faq_categories li a { font-size: 15px; line-height: 20px; display: inline-block; width: 120px; height: 60px; padding-top: 60px; border-radius: 50%; -moz-border-radius: 50%; -webkit-border-radius: 50%; text-align: center; /* padding: 10px 15px 10px 50px; */ background: #fff; -webkit-box-shadow: 0px 3px 3px rgba(0,0,0,0.25); -moz-box-shadow: 0px 3px 3px rgba(0,0,0,0.25); box-shadow: 0px 3px 3px rgba(0,0,0,0.25); } @media only screen and (min-width: 992px){ .tcode-event-schedule .col-md-2 { width: 8.33333333%; } .tcode-event-schedule .col-md-9 { width: 83.33333333%; } .tcode-event-schedule .col-md-8{ width: 83.33333333%; } .tcode-event-schedule .col-md-offset-2 { margin-left: 8.33333333%; } } </style> <!-- end Simple Custom CSS and JS --> <!-- DO NOT COPY THIS SNIPPET! Start of Page Analytics Tracking for HubSpot WordPress plugin v11.1.66--> <script class="hsq-set-content-id" data-content-id="standard-page"> var _hsq = _hsq || []; _hsq.push(["setContentType", "standard-page"]); </script> <!-- DO NOT COPY THIS SNIPPET! End of Page Analytics Tracking for HubSpot WordPress plugin --> <noscript><style type="text/css">.mptt-shortcode-wrapper .mptt-shortcode-table:first-of-type{display:table!important}.mptt-shortcode-wrapper .mptt-shortcode-table .mptt-event-container:hover{height:auto!important;min-height:100%!important}body.mprm_ie .mptt-shortcode-wrapper .mptt-event-container{height:auto!important}@media (max-width:767px){.mptt-shortcode-wrapper .mptt-shortcode-table:first-of-type{display:none!important}}</style></noscript><meta name="facebook-domain-verification" content="am2vglo5kisnxesykxjg911bcapxcr" /> <style> .schedule-tab-wrapper .etn-nav li a { font-size: 1.25rem; font-weight: 700; color: #ffffff; text-align: center; } .schedule-tab-wrapper .etn-nav li { /*display: none;*/ margin: 0; } .row-day span{font-size:20px;} </style> <style> .ui-tabs .ui-tabs-nav li.ui-tabs-active a { border-bottom: 2px solid; color: white!important; font-size: 30px; } .ui-tabs .ui-tabs-nav li a{ border-bottom: 2px solid; color: #1a9fe5; font-size: 30px;} .ui-tabs .ui-tabs-nav { margin: 0 auto; padding: 0; display: table; } /*.etn-date{display:none!important}*/ </style> <script type="text/javascript"> var ajaxurl = 'https://odsc.com/wp-admin/admin-ajax.php'; </script> <script type='text/javascript'> var video_popup_unprm_general_settings = { 'unprm_r_border': '' }; </script> <style type="text/css" media="all" id="wcs_styles"></style><noscript><style>.wp-grid-builder .wpgb-card.wpgb-card-hidden .wpgb-card-wrapper{opacity:1!important;visibility:visible!important;transform:none!important}.wpgb-facet {opacity:1!important;pointer-events:auto!important}.wpgb-facet *:not(.wpgb-pagination-facet){display:none}</style></noscript> <!-- Google Tag Manager for WordPress by gtm4wp.com --> <!-- GTM Container placement set to footer --> <script data-cfasync="false" data-pagespeed-no-defer> var dataLayer_content = {"pagePostType":"page","pagePostType2":"single-page","pagePostAuthor":"Kevin M"}; dataLayer.push( dataLayer_content ); </script> <script data-cfasync="false"> (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start': new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0], j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src= '//www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f); })(window,document,'script','dataLayer','GTM-M62L34N'); </script> <!-- End Google Tag Manager for WordPress by gtm4wp.com --><style id = "be-dynamic-css" type="text/css"> body { background-color: rgb(255,255,255);background-color: rgba(255,255,255,1);} .layout-box #header-inner-wrap, #header-inner-wrap, body.header-transparent #header #header-inner-wrap.no-transparent, .left-header .sb-slidebar.sb-left, .left-header .sb-slidebar.sb-left #slidebar-menu a::before { background-color: rgb(50,56,64);background-color: rgba(50,56,64,1);} #mobile-menu, #mobile-menu ul { background-color: rgb(50,56,64);background-color: rgba(50,56,64,1);} #mobile-menu li{ border-bottom-color: #efefef ; } body.header-transparent #header-inner-wrap{ background: transparent; } .be-gdpr-modal-item input:checked + .slider{ background-color: #00bcdd; } .be-gdpr-modal-iteminput:focus + .slider { box-shadow: 0 0 1px #00bcdd; } .be-gdpr-modal-item .slider:before { background-color:#ffffff; } .be-gdpr-cookie-notice-bar .be-gdpr-cookie-notice-button{ background: #00bcdd; color: #ffffff; } #header .header-border{ border-bottom: 0px none #00bcdd; } #header-top-bar{ background-color: rgb(21,22,26);background-color: rgba(21,22,26,1); border-bottom: 0px none #ffffff; color: #ffffff; } #header-top-bar #topbar-menu li a{ color: #ffffff; } #header-bottom-bar{ background-color: rgb(255,255,255);background-color: rgba(255,255,255,1); border-top: none #323232; border-bottom: none #323232; } /*Adjusted the timings for the new effects*/ body.header-transparent #header #header-inner-wrap { -webkit-transition: background .25s ease, box-shadow .25s ease, opacity 700ms cubic-bezier(0.645, 0.045, 0.355, 1), transform 700ms cubic-bezier(0.645, 0.045, 0.355, 1); -moz-transition: background .25s ease, box-shadow .25s ease, opacity 700ms cubic-bezier(0.645, 0.045, 0.355, 1), transform 700ms cubic-bezier(0.645, 0.045, 0.355, 1); -o-transition: background .25s ease, box-shadow .25s ease, opacity 700ms cubic-bezier(0.645, 0.045, 0.355, 1), transform 700ms cubic-bezier(0.645, 0.045, 0.355, 1); transition: background .25s ease, box-shadow .25s ease, opacity 700ms cubic-bezier(0.645, 0.045, 0.355, 1), transform 700ms cubic-bezier(0.645, 0.045, 0.355, 1); } body.header-transparent.semi #header .semi-transparent{ background-color: rgb(255,255,255);background-color: rgba(255,255,255,1); !important ; } #content, #blog-content { background-color: rgb(255,255,255);background-color: rgba(255,255,255,1);} #bottom-widgets { background-color: rgb(242,243,248);background-color: rgba(242,243,248,1);} #footer { background-color: rgb(50,56,64);background-color: rgba(50,56,64,1);} #footer .footer-border{ border-bottom: 2px solid #eaeaea; } .page-title-module-custom { background-color: rgb(242,243,248);background-color: rgba(242,243,248,1);} #portfolio-title-nav-wrap{ background-color : #ededed; } #navigation .sub-menu, #navigation .children, #navigation-left-side .sub-menu, #navigation-left-side .children, #navigation-right-side .sub-menu, #navigation-right-side .children { background-color: rgb(31,31,31);background-color: rgba(31,31,31,1);} .sb-slidebar.sb-right { background-color: rgb(26,26,26);background-color: rgba(26,26,26,1);} .left-header .left-strip-wrapper, .left-header #left-header-mobile { background-color : #323840 ; } .layout-box-top, .layout-box-bottom, .layout-box-right, .layout-box-left, .layout-border-header-top #header-inner-wrap, .layout-border-header-top.layout-box #header-inner-wrap, body.header-transparent .layout-border-header-top #header #header-inner-wrap.no-transparent { background-color: rgb(211,211,211);background-color: rgba(211,211,211,1);} .left-header.left-sliding.left-overlay-menu .sb-slidebar{ background-color: rgb(8,8,8);background-color: rgba(8,8,8,0.90); } .top-header.top-overlay-menu .sb-slidebar{ background-color: rgb(26,26,26);background-color: rgba(26,26,26,1);} .search-box-wrapper{ background-color: rgb(255,255,255);background-color: rgba(255,255,255,0.85);} .search-box-wrapper.style1-header-search-widget input[type="text"]{ background-color: transparent !important; color: #000000; border: 1px solid #000000; } .search-box-wrapper.style2-header-search-widget input[type="text"]{ background-color: transparent !important; color: #000000; border: none !important; box-shadow: none !important; } .search-box-wrapper .searchform .search-icon{ color: #000000; } #header-top-bar-right .search-box-wrapper.style1-header-search-widget input[type="text"]{ border: none; } .post-title , .post-date-wrap { margin-bottom: 12px; } /* ====================== Dynamic Border Styling ====================== */ .layout-box-top, .layout-box-bottom { height: 30px; } .layout-box-right, .layout-box-left { width: 30px; } #main.layout-border, #main.layout-border.layout-border-header-top{ padding: 30px; } .left-header #main.layout-border { padding-left: 0px; } #main.layout-border.layout-border-header-top { padding-top: 0px; } .be-themes-layout-layout-border #logo-sidebar, .be-themes-layout-layout-border-header-top #logo-sidebar{ margin-top: 70px; } /*Left Static Menu*/ .left-header.left-static.be-themes-layout-layout-border #main-wrapper{ margin-left: 310px; } .left-header.left-static.be-themes-layout-layout-border .sb-slidebar.sb-left { left: 30px; } /*Right Slidebar*/ body.be-themes-layout-layout-border-header-top .sb-slidebar.sb-right, body.be-themes-layout-layout-border .sb-slidebar.sb-right { right: -250px; } .be-themes-layout-layout-border-header-top .sb-slidebar.sb-right.opened, .be-themes-layout-layout-border .sb-slidebar.sb-right.opened { right: 30px; } /* Top-overlay menu on opening, header moves sideways bug. Fixed on the next line code */ /*body.be-themes-layout-layout-border-header-top.top-header.slider-bar-opened #main #header #header-inner-wrap.no-transparent.top-animate, body.be-themes-layout-layout-border.top-header.slider-bar-opened #main #header #header-inner-wrap.no-transparent.top-animate { right: 310px; }*/ body.be-themes-layout-layout-border-header-top.top-header:not(.top-overlay-menu).slider-bar-opened #main #header #header-inner-wrap.no-transparent.top-animate, body.be-themes-layout-layout-border.top-header:not(.top-overlay-menu).slider-bar-opened #main #header #header-inner-wrap.no-transparent.top-animate { right: 310px; } /* Now not needed mostly, as the hero section image is coming properly */ /*Single Page Version*/ body.be-themes-layout-layout-border-header-top.single-page-version .single-page-nav-wrap, body.be-themes-layout-layout-border.single-page-version .single-page-nav-wrap { right: 50px; } /*Split Screen Page Template*/ .top-header .layout-border #content.page-split-screen-left { margin-left: calc(50% + 15px); } .top-header.page-template-page-splitscreen-left .layout-border .header-hero-section { width: calc(50% - 15px); } .top-header .layout-border #content.page-split-screen-right { width: calc(50% - 15px); } .top-header.page-template-page-splitscreen-right .layout-border .header-hero-section { left: calc(50% - 15px); } @media only screen and (max-width: 960px) { body.be-themes-layout-layout-border-header-top.single-page-version .single-page-nav-wrap, body.be-themes-layout-layout-border.single-page-version .single-page-nav-wrap { right: 35px; } body.be-themes-layout-layout-border-header-top .sb-slidebar.sb-right, body.be-themes-layout-layout-border .sb-slidebar.sb-right { right: -280px; } #main.layout-border, #main.layout-border.layout-border-header-top { padding: 0px !important; } .top-header .layout-border #content.page-split-screen-left, .top-header .layout-border #content.page-split-screen-right { margin-left: 0px; width:100%; } .top-header.page-template-page-splitscreen-right .layout-border .header-hero-section, .top-header.page-template-page-splitscreen-left .layout-border .header-hero-section { width:100%; } } body, .special-heading-wrap .caption-wrap .body-font, .woocommerce .woocommerce-ordering select.orderby, .woocommerce-page .woocommerce-ordering select.orderby { font: 400 18px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #5f6263; line-height: 26px; letter-spacing: 0px; text-transform: none; -webkit-font-smoothing: antialiased; -moz-osx-font-smoothing: grayscale; } h1 { font: 700 55px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #515151; line-height: 70px; letter-spacing: 0px; text-transform: none;} h2 { font: 700 42px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #515151; line-height: 63px; letter-spacing: 0px; text-transform: none;} h3 { font: 700 35px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #515151; line-height: 52px; letter-spacing: 1px; text-transform: none;} h4, .woocommerce-order-received .woocommerce h2, .woocommerce-order-received .woocommerce h3, .woocommerce-view-order .woocommerce h2, .woocommerce-view-order .woocommerce h3{ font: 400 26px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #515151; line-height: 42px; letter-spacing: 0px; text-transform: none;} h5, #reply-title { font: 400 20px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #515151; line-height: 36px; letter-spacing: 0px; text-transform: none; } h6, .testimonial-author-role.h6-font, .menu-card-title, .menu-card-item-price, .slider-counts, .woocommerce-MyAccount-navigation ul li { font: 400 15px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #515151; line-height: 32px; letter-spacing: 0px; text-transform: none;} h6.gallery-side-heading { font-size: 18px; } .special-subtitle , .style1.thumb-title-wrap .portfolio-item-cats { font-style: ; font-size: 15px; font-weight: ; font-family: Crimson Text; text-transform: none; letter-spacing: 0px; } .gallery-side-heading { font-size: 18px; } .attachment-details-custom-slider { background-color: rgb(0,0,0);background-color: rgba(0,0,0,1); font: 15px "Crimson Text","Open Sans","Arial",sans-serif; color: ; line-height: 15px; letter-spacing: 0px; text-transform: none;} .single-portfolio-slider .carousel_bar_wrap { background-color: rgb(255,255,255);background-color: rgba(255,255,255,0.5);} .top-right-sliding-menu .sb-right ul#slidebar-menu li, .overlay-menu-close, .be-overlay-menu-close { font: 400 16px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #ffffff; line-height: 50px; letter-spacing: 1px; text-transform: uppercase;} .top-right-sliding-menu .sb-right ul#slidebar-menu li a { color: #ffffff !important; } .top-right-sliding-menu .sb-right #slidebar-menu ul.sub-menu li { font: 400 16px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #ffffff; line-height: 25px; letter-spacing: ; text-transform: none;} .top-right-sliding-menu .sb-right ul#slidebar-menu li a { color: #ffffff !important; } .sb-right #slidebar-menu .mega .sub-menu .highlight .sf-with-ul { font: 400 16px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #ffffff; line-height: 50px; letter-spacing: 1px; text-transform: uppercase; color: #ffffff !important; } .post-meta.post-top-meta-typo, .style8-blog .post-meta.post-category a, .hero-section-blog-categories-wrap a { font: 12px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #757575; line-height: 24px; letter-spacing: 0px; text-transform: uppercase;; } #portfolio-title-nav-bottom-wrap h6, #portfolio-title-nav-bottom-wrap .slider-counts { font: 400 15px "Montserrat","Open Sans","Arial",sans-serif; color: ; line-height: ; letter-spacing: 0px; text-transform: none;; line-height: 40px; } .filters .filter_item { font: 400 12px "Montserrat","Open Sans","Arial",sans-serif; color: #222222; line-height: 32px; letter-spacing: 1px; text-transform: uppercase;; } ul#mobile-menu a { font: 400 12px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #ffffff; line-height: 40px; letter-spacing: 1px; text-transform: uppercase;} ul#mobile-menu ul.sub-menu a { font: 400 13px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #bbbbbb; line-height: 27px; letter-spacing: 0px; text-transform: none; } ul#mobile-menu li.mega ul.sub-menu li.highlight > :first-child { font: 400 12px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #ffffff; line-height: 40px; letter-spacing: 1px; text-transform: uppercase;} #navigation, .style2 #navigation, .style13 #navigation, #navigation-left-side, #navigation-right-side, .sb-left #slidebar-menu, .header-widgets, .header-code-widgets, body #header-inner-wrap.top-animate.style2 #navigation, .top-overlay-menu .sb-right #slidebar-menu, #navigation .mega .sub-menu .highlight .sf-with-ul, .special-header-menu .menu-container { font: 600 16px "Montserrat","Open Sans","Arial",sans-serif; color: #00bcdd; line-height: 51px; letter-spacing: 1px; text-transform: uppercase;} #navigation .sub-menu, #navigation .children, #navigation-left-side .sub-menu, #navigation-left-side .children, #navigation-right-side .sub-menu, #navigation-right-side .children, .sb-left #slidebar-menu .sub-menu, .top-overlay-menu .sb-right #slidebar-menu .sub-menu, .special-header-menu .menu-container .sub-menu, .special-header-menu .sub-menu { font: 400 16px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #bbbbbb; line-height: 28px; letter-spacing: 0px; text-transform: none;} .thumb-title-wrap .thumb-title { font: 400 14px "Montserrat","Open Sans","Arial",sans-serif; color: ; line-height: 30px; letter-spacing: 0px; text-transform: uppercase;} .thumb-title-wrap .portfolio-item-cats { font-size: 12px; line-height: 17px; text-transform: none; letter-spacing: 0px; } .full-screen-portfolio-overlay-title { font: 400 14px "Montserrat","Open Sans","Arial",sans-serif; color: ; line-height: 30px; letter-spacing: 0px; text-transform: uppercase;} #footer { font: 400 14px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #ffffff; line-height: 14px; letter-spacing: 0px; text-transform: none;} #bottom-widgets h6 { font: 400 18px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #474747; line-height: 22px; letter-spacing: 1px; text-transform: uppercase; margin-bottom:20px; } #bottom-widgets { font: 400 16px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #757575; line-height: 24px; letter-spacing: 0px; text-transform: none;} .sidebar-widgets h6 { font: 400 18px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #333333; line-height: 22px; letter-spacing: 1px; text-transform: uppercase; margin-bottom:20px; } .sidebar-widgets { ?php be_themes_print_typography('sidebar_widget_text'); ?> } .sb-slidebar .widget { font: 400 14px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #a2a2a2; line-height: 25px; letter-spacing: 0px; text-transform: none;} .sb-slidebar .widget h6 { font: 400 16px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #ffffff; line-height: 22px; letter-spacing: 1px; text-transform: none;} .woocommerce ul.products li.product .product-meta-data h3, .woocommerce-page ul.products li.product .product-meta-data h3, .woocommerce ul.products li.product h3, .woocommerce-page ul.products li.product h3 { font: 400 13px "Montserrat","Open Sans","Arial",sans-serif; color: #222222; line-height: 27px; letter-spacing: 1px; text-transform: uppercase;} .woocommerce ul.products li.product .product-meta-data .woocommerce-loop-product__title, .woocommerce-page ul.products li.product .product-meta-data .woocommerce-loop-product__title, .woocommerce ul.products li.product .woocommerce-loop-product__title, .woocommerce-page ul.products li.product .woocommerce-loop-product__title, .woocommerce ul.products li.product-category .woocommerce-loop-category__title, .woocommerce-page ul.products li.product-category .woocommerce-loop-category__title { font: 400 13px "Montserrat","Open Sans","Arial",sans-serif; color: #222222; line-height: 27px; letter-spacing: 1px; text-transform: uppercase; margin-bottom:5px; text-align: center; } .woocommerce-page.single.single-product #content div.product h1.product_title.entry-title { font: 400 25px "Montserrat","Open Sans","Arial",sans-serif; color: #222222; line-height: 27px; letter-spacing: 0px; text-transform: none;} .contact_form_module input[type="text"], .contact_form_module textarea { font: 400 13px "Montserrat","Open Sans","Arial",sans-serif; color: #222222; line-height: 26px; letter-spacing: 0px; text-transform: none;} .page-title-module-custom .page-title-custom, h6.portfolio-title-nav{ font: 400 18px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #515151; line-height: 36px; letter-spacing: 3px; text-transform: uppercase;} .tatsu-button, .be-button, .woocommerce a.button, .woocommerce-page a.button, .woocommerce button.button, .woocommerce-page button.button, .woocommerce input.button, .woocommerce-page input.button, .woocommerce #respond input#submit, .woocommerce-page #respond input#submit, .woocommerce #content input.button, .woocommerce-page #content input.button, input[type="submit"], .more-link.style1-button, .more-link.style2-button, .more-link.style3-button, input[type="button"], input[type="submit"], input[type="reset"], button, input[type="file"]::-webkit-file-upload-button { font-family: Source Sans Pro; font-weight: ; } .post-title , .post-date-wrap { font: 400 20px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #000000; line-height: 40px; letter-spacing: 0px; text-transform: none; margin-bottom: 12px; } .style3-blog .post-title, .style8-blog .post-title { font: 400 16px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #363c3b; line-height: 28px; letter-spacing: 0px; text-transform: none;} .post-nav li, .style8-blog .post-meta.post-date, .style8-blog .post-bottom-meta-wrap, .hero-section-blog-bottom-meta-wrap { font: 12px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #757575; line-height: 24px; letter-spacing: 0px; text-transform: uppercase;} .ui-tabs-anchor, .accordion .accordion-head, .skill-wrap .skill_name, .chart-wrap span, .animate-number-wrap h6 span, .woocommerce-tabs .tabs li a, .be-countdown { font-family: Source Sans Pro; letter-spacing: 0px; font-style: ; font-weight: 600; } .ui-tabs-anchor { font-size: 13px; line-height: 17px; text-transform: uppercase; } .accordion .accordion-head { font-size: 18px; line-height: 17px; text-transform: uppercase; } .skill-wrap .skill_name { font-size: 20px; line-height: 17px; text-transform: none; } .countdown-section { font-size: 15px; line-height: 30px; text-transform: uppercase; } .countdown-amount { font-size: 55px; line-height: 95px; text-transform: uppercase; } .tweet-slides .tweet-content { font-family: Source Sans Pro; letter-spacing: 0px; font-style: ; font-weight: ; text-transform: none; } .testimonial_slide .testimonial-content { font-family: Crimson Text; letter-spacing: 0px; font-style: ; font-weight: ; text-transform: none; } .oshine-animated-link, .view-project-link.style4-button { font-family: Montserrat; font-weight: ; letter-spacing: ; font-style: ; text-transform: none; } a.navigation-previous-post-link, a.navigation-next-post-link { font: 700 13px "Montserrat","Open Sans","Arial",sans-serif; color: ; line-height: 20px; letter-spacing: 0px; text-transform: none;; } @media only screen and (max-width : 767px ) { #hero-section h1 , .full-screen-section-wrap h1, .tatsu-fullscreen-wrap h1 { font-size: 30px; line-height: 40px; } #hero-section h2, .full-screen-section-wrap h2, .tatsu-fullscreen-wrap h2 { font-size: 25px; line-height: 35px; } #hero-section h4, .full-screen-section-wrap h4, .tatsu-fullscreen-wrap h3 { font-size: 16px; line-height: 30px; } #hero-section h5, .full-screen-section-wrap h5, .tatsu-fullscreen-wrap h5 { font-size: 16px; line-height: 30px; } } /* RELATED TO TYPOGRAPHY */ #header-controls-right, #header-controls-left { color: #00bcdd} #be-left-strip .be-mobile-menu-icon span { background-color: #323232} ul#mobile-menu .mobile-sub-menu-controller { line-height : 40px ; } ul#mobile-menu ul.sub-menu .mobile-sub-menu-controller{ line-height : 27px ; } .breadcrumbs { color: #515151; } .search-box-wrapper.style2-header-search-widget input[type="text"]{ font-style: ; font-weight: ; font-family: Crimson Text; } .portfolio-share a.custom-share-button, .portfolio-share a.custom-share-button:active, .portfolio-share a.custom-share-button:hover, .portfolio-share a.custom-share-button:visited { color: #515151; } .more-link.style2-button { color: #000000 !important; border-color: #000000 !important; } .style8-blog .post-bottom-meta-wrap .be-share-stack a.custom-share-button, .style8-blog .post-bottom-meta-wrap .be-share-stack a.custom-share-button:active, .style8-blog .post-bottom-meta-wrap .be-share-stack a.custom-share-button:hover, .style8-blog .post-bottom-meta-wrap .be-share-stack a.custom-share-button:visited { color: #757575; } .hero-section-blog-categories-wrap a, .hero-section-blog-categories-wrap a:visited, .hero-section-blog-categories-wrap a:hover, .hero-section-blog-bottom-meta-wrap .hero-section-blog-bottom-meta-wrap a, .hero-section-blog-bottom-meta-wrap a:visited, .hero-section-blog-bottom-meta-wrap a:hover { color : #000000; } #navigation .mega .sub-menu .highlight .sf-with-ul { color: #bbbbbb !important; line-height:1.5; } .view-project-link.style4-button { color : #515151; } .pricing-table .pricing-feature{ font-size: 17px; } /* Woocommerce */ .related.products h2, .upsells.products h2, .cart-collaterals .cross-sells h2, .cart_totals h2, .shipping_calculator h2, .woocommerce-billing-fields h3, .woocommerce-shipping-fields h3, .shipping_calculator h2, #order_review_heading, .woocommerce .page-title { font-family: Montserrat; font-weight: 400; } .woocommerce form .form-row label, .woocommerce-page form .form-row label { color: #515151; } .woocommerce-tabs .tabs li a { color: #515151 !important; } /* BB Press Plugin */ a.bbp-forum-title, #bbpress-forums fieldset.bbp-form label, .bbp-topic-title a.bbp-topic-permalink { font: 400 15px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #515151; line-height: 32px; letter-spacing: 0px; text-transform: none;} #bbpress-forums ul.forum-titles li, #bbpress-forums ul.bbp-replies li.bbp-header { font: 400 15px "Source Sans Pro","Open Sans","Arial",sans-serif; color: #515151; line-height: 32px; letter-spacing: 0px; text-transform: none; line-height: inherit; letter-spacing: inherit; text-transform: uppercase; font-size: inherit; } #bbpress-forums .topic .bbp-topic-meta a, .bbp-forum-freshness a, .bbp-topic-freshness a, .bbp-header .bbp-reply-content a, .bbp-topic-tags a, .bbp-breadcrumb a, .bbp-forums-list a { color: #515151; } /*Event On Plugin*/ .ajde_evcal_calendar .calendar_header p, .eventon_events_list .eventon_list_event .evcal_cblock { font-family: Source Sans Pro !important; } .eventon_events_list .eventon_list_event .evcal_desc span.evcal_desc2, .evo_pop_body .evcal_desc span.evcal_desc2 { font-family: Source Sans Pro !important; font-size: 14px !important; text-transform: none; } .eventon_events_list .eventon_list_event .evcal_desc span.evcal_event_subtitle, .evo_pop_body .evcal_desc span.evcal_event_subtitle, .evcal_evdata_row .evcal_evdata_cell p, #evcal_list .eventon_list_event p.no_events { text-transform: none !important; font-family: Source Sans Pro !important; font-size: inherit !important; } /* END RELATED TO TYPOGRAPHY */ .filters.single_border .filter_item{ border-color: #00bcdd; } .filters.rounded .current_choice{ border-radius: 50px; background-color: #00bcdd; color: #ffffff; } .filters.single_border .current_choice, .filters.border .current_choice{ color: #00bcdd; } .exclusive-mobile-bg .menu-controls{ background-color: background-color: rgb(255,255,255);background-color: rgba(255,255,255,0);; } #header .be-mobile-menu-icon span { background-color : #323232; } #header-controls-right, #header-controls-left, .overlay-menu-close, .be-overlay-menu-close { color : #323232; } #header .exclusive-mobile-bg .be-mobile-menu-icon, #header .exclusive-mobile-bg .be-mobile-menu-icon span, #header-inner-wrap.background--light.transparent.exclusive-mobile-bg .be-mobile-menu-icon, #header-inner-wrap.background--light.transparent.exclusive-mobile-bg .be-mobile-menu-icon span, #header-inner-wrap.background--dark.transparent.exclusive-mobile-bg .be-mobile-menu-icon, #header-inner-wrap.background--dark.transparent.exclusive-mobile-bg .be-mobile-menu-icon span { background-color: #323232} .be-mobile-menu-icon{ width: 18px; height: 2px; } .be-mobile-menu-icon .hamburger-line-1{ top: -5px; } .be-mobile-menu-icon .hamburger-line-3{ top: 5px; } .thumb-title-wrap { color: #ffffff; } #bottom-widgets .widget ul li a, #bottom-widgets a { color: inherit; } #bottom-widgets .tagcloud a:hover { color: #ffffff; } a, a:visited, a:hover, #bottom-widgets .widget ul li a:hover, #bottom-widgets a:hover{ color: #00bcdd; } #header-top-menu a:hover, #navigation .current_page_item a, #navigation .current_page_item a:hover, #navigation a:hover, #navigation-left-side .current_page_item a, #navigation-left-side .current_page_item a:hover, #navigation-left-side a:hover, #navigation-right-side .current_page_item a, #navigation-right-side .current_page_item a:hover, #navigation-right-side a:hover, #menu li.current-menu-ancestor > a, #navigation .current-menu-item > a, #navigation .sub-menu .current-menu-item > a, #navigation .sub-menu a:hover, #navigation .children .current-menu-item > a, #navigation .children a:hover, #slidebar-menu .current-menu-item > a, .special-header-menu a:hover + .mobile-sub-menu-controller i, .special-header-menu #slidebar-menu a:hover, .special-header-menu .sub-menu a:hover, .single-page-version #navigation a:hover, .single-page-version #navigation-left-side a:hover, .single-page-version #navigation-right-side a:hover, .single-page-version #navigation .current-section.current_page_item a, .single-page-version #navigation-left-side .current-section.current_page_item a, .single-page-version #navigation-right-side .current-section.current_page_item a, .single-page-version #slidebar-menu .current-section.current_page_item a, .single-page-version #navigation .current_page_item a:hover, .single-page-version #navigation-left-side .current_page_item a:hover, .single-page-version #navigation-right-side .current_page_item a:hover, .single-page-version #slidebar-menu .current_page_item a:hover, .be-sticky-sections #navigation a:hover, .be-sticky-sections #navigation-left-side a:hover, .be-sticky-sections #navigation-right-side a:hover, .be-sticky-sections #navigation .current-section.current_page_item a, .be-sticky-sections #navigation-left-side .current-section.current_page_item a, .be-sticky-sections #navigation-right-side .current-section.current_page_item a, .be-sticky-sections #navigation .current_page_item a:hover, .be-sticky-sections #navigation-left-side .current_page_item a:hover, .be-sticky-sections #navigation-right-side .current_page_item a:hover, #navigation .current-menu-ancestor > a, #navigation-left-side .current-menu-ancestor > a, #navigation-right-side .current-menu-ancestor > a, #slidebar-menu .current-menu-ancestor > a, .special-header-menu .current-menu-item > a, .sb-left #slidebar-menu a:hover { color: #ffffff; } #navigation .current_page_item ul li a, #navigation-left-side .current_page_item ul li a, #navigation-right-side .current_page_item ul li a, .single-page-version #navigation .current_page_item a, .single-page-version #navigation-left-side .current_page_item a, .single-page-version #navigation-right-side .current_page_item a, .single-page-version #slidebar-menu .current_page_item a, .single-page-version #navigation .sub-menu .current-menu-item > a, .single-page-version #navigation .children .current-menu-item > a .be-sticky-sections #navigation .current_page_item a, .be-sticky-sections #navigation-left-side .current_page_item a, .be-sticky-sections #navigation-right-side .current_page_item a, .be-sticky-sections #navigation .sub-menu .current-menu-item > a, .be-sticky-sections #navigation .children .current-menu-item > a { color: inherit; } .be-nav-link-effect-1 a::after, .be-nav-link-effect-2 a::after, .be-nav-link-effect-3 a::after{ background-color: rgb(255,255,255);background-color: rgba(255,255,255,1);} #portfolio-title-nav-wrap .portfolio-nav a { color: #00bcdd; } #portfolio-title-nav-wrap .portfolio-nav a .home-grid-icon span{ background-color: #00bcdd; } #portfolio-title-nav-wrap .portfolio-nav a:hover { color: #323840; } #portfolio-title-nav-wrap .portfolio-nav a:hover .home-grid-icon span{ background-color: #323840; } .page-title-module-custom .header-breadcrumb { line-height: 36px; } #portfolio-title-nav-bottom-wrap h6, #portfolio-title-nav-bottom-wrap ul li a, .single_portfolio_info_close, #portfolio-title-nav-bottom-wrap .slider-counts{ background-color: rgb(255,255,255);background-color: rgba(255,255,255,0);} .more-link.style2-button:hover { border-color: #00bcdd !important; background: #00bcdd !important; color: #ffffff !important; } .woocommerce a.button, .woocommerce-page a.button, .woocommerce button.button, .woocommerce-page button.button, .woocommerce input.button, .woocommerce-page input.button, .woocommerce #respond input#submit, .woocommerce-page #respond input#submit, .woocommerce #content input.button, .woocommerce-page #content input.button { background: transparent !important; color: #000 !important; border-color: #000 !important; border-style: solid !important; border-width: 2px !important; background: transparent !important; color: #000000 !important; border-width: 2px !important; border-color: #000000 !important; line-height: 41px; text-transform: uppercase; } .woocommerce a.button:hover, .woocommerce-page a.button:hover, .woocommerce button.button:hover, .woocommerce-page button.button:hover, .woocommerce input.button:hover, .woocommerce-page input.button:hover, .woocommerce #respond input#submit:hover, .woocommerce-page #respond input#submit:hover, .woocommerce #content input.button:hover, .woocommerce-page #content input.button:hover { background: #e0a240 !important; color: #fff !important; border-color: #e0a240 !important; border-width: 2px !important; background: #e0a240 !important; color: #ffffff !important; border-color: #e0a240 !important; } .woocommerce a.button.alt, .woocommerce-page a.button.alt, .woocommerce .button.alt, .woocommerce-page .button.alt, .woocommerce input.button.alt, .woocommerce-page input.button.alt, .woocommerce input[type="submit"].alt, .woocommerce-page input[type="submit"].alt, .woocommerce #respond input#submit.alt, .woocommerce-page #respond input#submit.alt, .woocommerce #content input.button.alt, .woocommerce-page #content input.button.alt { background: #e0a240 !important; color: #fff !important; border-color: #e0a240 !important; border-style: solid !important; border-width: 2px !important; background: #e0a240 !important; color: #ffffff !important; border-width: 2px !important; border-color: #e0a240 !important; line-height: 41px; text-transform: uppercase; } .woocommerce a.button.alt:hover, .woocommerce-page a.button.alt:hover, .woocommerce .button.alt:hover, .woocommerce-page .button.alt:hover, .woocommerce input[type="submit"].alt:hover, .woocommerce-page input[type="submit"].alt:hover, .woocommerce input.button.alt:hover, .woocommerce-page input.button.alt:hover, .woocommerce #respond input#submit.alt:hover, .woocommerce-page #respond input#submit.alt:hover, .woocommerce #content input.button.alt:hover, .woocommerce-page #content input.button.alt:hover { background: transparent !important; color: #000 !important; border-color: #000 !important; border-style: solid !important; border-width: 2px !important; background: transparent !important; color: #000000 !important; border-color: #000000 !important; } .woocommerce .woocommerce-message a.button, .woocommerce-page .woocommerce-message a.button, .woocommerce .woocommerce-message a.button:hover, .woocommerce-page .woocommerce-message a.button:hover { border: none !important; color: #fff !important; background: none !important; } .woocommerce .woocommerce-ordering select.orderby, .woocommerce-page .woocommerce-ordering select.orderby { border-color: #eeeeee; } .style7-blog .post-title{ margin-bottom: 9px; } .style8-blog .post-comment-wrap a:hover{ color : #00bcdd; } .style8-blog .element:not(.be-image-post) .post-details-wrap{ background-color: #ffffff ; } .accordion .accordion-head.with-bg.ui-accordion-header-active{ background-color: #00bcdd; color: #ffffff !important; } #portfolio-title-nav-wrap{ padding-top: 15px; padding-bottom: 15px; border-bottom: 1px solid #cecece; } #portfolio-title-nav-bottom-wrap h6, #portfolio-title-nav-bottom-wrap ul, .single_portfolio_info_close .font-icon, .slider-counts{ color: #2b2b2b ; } #portfolio-title-nav-bottom-wrap .home-grid-icon span{ background-color: #2b2b2b ; } #portfolio-title-nav-bottom-wrap h6:hover, #portfolio-title-nav-bottom-wrap ul a:hover, #portfolio-title-nav-bottom-wrap .slider-counts:hover, .single_portfolio_info_close:hover { background-color: rgb(235,73,73);background-color: rgba(235,73,73,0.85);} #portfolio-title-nav-bottom-wrap h6:hover, #portfolio-title-nav-bottom-wrap ul a:hover, #portfolio-title-nav-bottom-wrap .slider-counts:hover, .single_portfolio_info_close:hover .font-icon{ color: #ffffff ; } #portfolio-title-nav-bottom-wrap ul a:hover .home-grid-icon span{ background-color: #ffffff ; } /* ====================== Layout ====================== */ body #header-inner-wrap.top-animate #navigation, body #header-inner-wrap.top-animate .header-controls, body #header-inner-wrap.stuck #navigation, body #header-inner-wrap.stuck .header-controls { -webkit-transition: line-height 0.5s ease; -moz-transition: line-height 0.5s ease; -ms-transition: line-height 0.5s ease; -o-transition: line-height 0.5s ease; transition: line-height 0.5s ease; } .header-cart-controls .cart-contents span{ background: #646464; } .header-cart-controls .cart-contents span{ color: #f5f5f5; } .left-sidebar-page, .right-sidebar-page, .no-sidebar-page .be-section-pad:first-child, .page-template-page-940-php #content , .no-sidebar-page #content-wrap, .portfolio-archives.no-sidebar-page #content-wrap { padding-top: 80px; padding-bottom: 80px; } .no-sidebar-page #content-wrap.page-builder{ padding-top: 0px; padding-bottom: 0px; } .left-sidebar-page .be-section:first-child, .right-sidebar-page .be-section:first-child, .dual-sidebar-page .be-section:first-child { padding-top: 0 !important; } .style1 .logo, .style4 .logo, #left-header-mobile .logo, .style3 .logo, .style7 .logo, .style10 .logo{ padding-top: 15px; padding-bottom: 15px; } .style5 .logo, .style6 .logo{ margin-top: 15px; margin-bottom: 15px; } #footer-wrap { padding-top: 25px; padding-bottom: 25px; } /* ====================== Colors ====================== */ .sec-bg, .gallery_content, .fixed-sidebar-page .fixed-sidebar, .style3-blog .blog-post.element .element-inner, .style4-blog .blog-post, .blog-post.format-link .element-inner, .blog-post.format-quote .element-inner, .woocommerce ul.products li.product, .woocommerce-page ul.products li.product, .chosen-container.chosen-container-single .chosen-drop, .chosen-container.chosen-container-single .chosen-single, .chosen-container.chosen-container-active.chosen-with-drop .chosen-single { background: #fafbfd; } .sec-color, .post-meta a, .pagination a, .pagination a:visited, .pagination span, .pages_list a, input[type="text"], input[type="email"], input[type="password"], textarea, .gallery_content, .fixed-sidebar-page .fixed-sidebar, .style3-blog .blog-post.element .element-inner, .style4-blog .blog-post, .blog-post.format-link .element-inner, .blog-post.format-quote .element-inner, .woocommerce ul.products li.product, .woocommerce-page ul.products li.product, .chosen-container.chosen-container-single .chosen-drop, .chosen-container.chosen-container-single .chosen-single, .chosen-container.chosen-container-active.chosen-with-drop .chosen-single { color: #7a7a7a; } .woocommerce .quantity .plus, .woocommerce .quantity .minus, .woocommerce #content .quantity .plus, .woocommerce #content .quantity .minus, .woocommerce-page .quantity .plus, .woocommerce-page .quantity .minus, .woocommerce-page #content .quantity .plus, .woocommerce-page #content .quantity .minus, .woocommerce .quantity input.qty, .woocommerce #content .quantity input.qty, .woocommerce-page .quantity input.qty, .woocommerce-page #content .quantity input.qty { background: #fafbfd; color: #7a7a7a; border-color: #eeeeee; } .woocommerce div.product .woocommerce-tabs ul.tabs li, .woocommerce #content div.product .woocommerce-tabs ul.tabs li, .woocommerce-page div.product .woocommerce-tabs ul.tabs li, .woocommerce-page #content div.product .woocommerce-tabs ul.tabs li { color: #7a7a7a!important; } .chosen-container .chosen-drop, nav.woocommerce-pagination, .summary.entry-summary .price, .portfolio-details.style2 .gallery-side-heading-wrap, #single-author-info, .single-page-atts, article.comment { border-color: #eeeeee !important; } .fixed-sidebar-page #page-content{ background: #ffffff; } .sec-border, input[type="text"], input[type="email"], input[type="tel"], input[type="password"], textarea { border: 2px solid #eeeeee; } .chosen-container.chosen-container-single .chosen-single, .chosen-container.chosen-container-active.chosen-with-drop .chosen-single { border: 2px solid #eeeeee; } .woocommerce table.shop_attributes th, .woocommerce-page table.shop_attributes th, .woocommerce table.shop_attributes td, .woocommerce-page table.shop_attributes td { border: none; border-bottom: 1px solid #eeeeee; padding-bottom: 5px; } .woocommerce .widget_price_filter .price_slider_wrapper .ui-widget-content, .woocommerce-page .widget_price_filter .price_slider_wrapper .ui-widget-content{ border: 1px solid #eeeeee; } .pricing-table .pricing-title, .chosen-container .chosen-results li { border-bottom: 1px solid #eeeeee; } .separator { border:0; height:1px; color: #eeeeee; background-color: #eeeeee; } .alt-color, li.ui-tabs-active h6 a, a, a:visited, .social_media_icons a:hover, .post-title a:hover, .fn a:hover, a.team_icons:hover, .recent-post-title a:hover, .widget_nav_menu ul li.current-menu-item a, .widget_nav_menu ul li.current-menu-item:before, .woocommerce ul.cart_list li a:hover, .woocommerce ul.product_list_widget li a:hover, .woocommerce-page ul.cart_list li a:hover, .woocommerce-page ul.product_list_widget li a:hover, .woocommerce-page .product-categories li a:hover, .woocommerce ul.products li.product .product-meta-data h3:hover, .woocommerce table.cart a.remove:hover, .woocommerce #content table.cart a.remove:hover, .woocommerce-page table.cart a.remove:hover, .woocommerce-page #content table.cart a.remove:hover, td.product-name a:hover, .woocommerce-page #content .quantity .plus:hover, .woocommerce-page #content .quantity .minus:hover, .post-category a:hover, a.custom-like-button.liked, .menu-card-item-stared { color: #00bcdd; } .content-slide-wrap .flex-control-paging li a.flex-active, .content-slide-wrap .flex-control-paging li.flex-active a:before { background: #00bcdd !important; border-color: #00bcdd !important; } #navigation .menu > ul > li.mega > ul > li { border-color: #3d3d3d; } .sb-slidebar.sb-right .menu{ border-top: 1px solid #2d2d2d; border-bottom: 1px solid #2d2d2d; } .post-title a:hover { color: #00bcdd !important; } .alt-bg, input[type="submit"], .tagcloud a:hover, .pagination a:hover, .widget_tag_cloud a:hover, .pagination .current, .trigger_load_more .be-button, .trigger_load_more .be-button:hover { background-color: #00bcdd; transition: 0.2s linear all; } .mejs-controls .mejs-time-rail .mejs-time-current , .mejs-controls .mejs-horizontal-volume-slider .mejs-horizontal-volume-current, .woocommerce span.onsale, .woocommerce-page span.onsale, .woocommerce a.add_to_cart_button.button.product_type_simple.added, .woocommerce-page .widget_shopping_cart_content .buttons a.button:hover, .woocommerce nav.woocommerce-pagination ul li span.current, .woocommerce nav.woocommerce-pagination ul li a:hover, .woocommerce nav.woocommerce-pagination ul li a:focus, .testimonial-flex-slider .flex-control-paging li a.flex-active, #back-to-top, .be-carousel-nav, .portfolio-carousel .owl-controls .owl-prev:hover, .portfolio-carousel .owl-controls .owl-next:hover, .owl-theme .owl-controls .owl-dot.active span, .owl-theme .owl-controls .owl-dot:hover span, .more-link.style3-button, .view-project-link.style3-button{ background: #00bcdd !important; } .single-page-nav-link.current-section-nav-link { background: #ffffff !important; } .view-project-link.style2-button, .single-page-nav-link.current-section-nav-link { border-color: #00bcdd !important; } .view-project-link.style2-button:hover { background: #00bcdd !important; color: #ffffff !important; } .tagcloud a:hover, .testimonial-flex-slider .flex-control-paging li a.flex-active, .testimonial-flex-slider .flex-control-paging li a { border-color: #00bcdd; } a.be-button.view-project-link, .more-link { border-color: #00bcdd; } .portfolio-container .thumb-bg { background-color: rgba(0,188,221,0.85); } .photostream_overlay, .be-button, .more-link.style3-button, .view-project-link.style3-button, button, input[type="button"], input[type="submit"], input[type="reset"] { background-color: #00bcdd; } input[type="file"]::-webkit-file-upload-button{ background-color: #00bcdd; } .alt-bg-text-color, input[type="submit"], .tagcloud a:hover, .pagination a:hover, .widget_tag_cloud a:hover, .pagination .current, .woocommerce nav.woocommerce-pagination ul li span.current, .woocommerce nav.woocommerce-pagination ul li a:hover, .woocommerce nav.woocommerce-pagination ul li a:focus, #back-to-top, .be-carousel-nav, .single_portfolio_close .font-icon, .single_portfolio_back .font-icon, .more-link.style3-button, .view-project-link.style3-button, .trigger_load_more a.be-button, .trigger_load_more a.be-button:hover, .portfolio-carousel .owl-controls .owl-prev:hover .font-icon, .portfolio-carousel .owl-controls .owl-next:hover .font-icon{ color: #ffffff; transition: 0.2s linear all; } .woocommerce .button.alt.disabled { background: #efefef !important; color: #a2a2a2 !important; border: none !important; cursor: not-allowed; } .be-button, input[type="button"], input[type="submit"], input[type="reset"], button { color: #ffffff; transition: 0.2s linear all; } input[type="file"]::-webkit-file-upload-button { color: #ffffff; transition: 0.2s linear all; } .button-shape-rounded #submit, .button-shape-rounded .style2-button.view-project-link, .button-shape-rounded .style3-button.view-project-link, .button-shape-rounded .style2-button.more-link, .button-shape-rounded .style3-button.more-link, .button-shape-rounded .contact_submit { border-radius: 3px; } .button-shape-circular .style2-button.view-project-link, .button-shape-circular .style3-button.view-project-link{ border-radius: 50px; padding: 17px 30px !important; } .button-shape-circular .style2-button.more-link, .button-shape-circular .style3-button.more-link{ border-radius: 50px; padding: 7px 30px !important; } .button-shape-circular .contact_submit, .button-shape-circular #submit{ border-radius: 50px; padding-left: 30px; padding-right: 30px; } .view-project-link.style4-button:hover::after{ border-color : #00bcdd; } .mfp-arrow{ color: #ffffff; transition: 0.2s linear all; -moz-transition: 0.2s linear all; -o-transition: 0.2s linear all; transition: 0.2s linear all; } .portfolio-title a { color: inherit; } .arrow-block .arrow_prev, .arrow-block .arrow_next, .arrow-block .flickity-prev-next-button { background-color: rgb(0,0,0);background-color: rgba(0,0,0,1);} .arrow-border .arrow_prev, .arrow-border .arrow_next, .arrow-border .flickity-prev-next-button { border: 1px solid #000000; } .gallery-info-box-wrap .arrow_prev .font-icon, .gallery-info-box-wrap .arrow_next .font-icon{ color: #ffffff; } .flickity-prev-next-button .arrow{ fill: #ffffff; } .arrow-block .arrow_prev:hover, .arrow-block .arrow_next:hover, .arrow-block .flickity-prev-next-button:hover { background-color: rgb(0,0,0);background-color: rgba(0,0,0,1);} .arrow-border .arrow_prev:hover, .arrow-border .arrow_next:hover, .arrow-border .flickity-prev-next-button:hover { border: 1px solid #000000; } .gallery-info-box-wrap .arrow_prev:hover .font-icon, .gallery-info-box-wrap .arrow_next:hover .font-icon{ color: #ffffff; } .flickity-prev-next-button:hover .arrow{ fill: #ffffff; } #back-to-top.layout-border, #back-to-top.layout-border-header-top { right: 50px; bottom: 50px; } .layout-border .fixed-sidebar-page #right-sidebar.active-fixed { right: 30px; } body.header-transparent.admin-bar .layout-border #header #header-inner-wrap.no-transparent.top-animate, body.sticky-header.admin-bar .layout-border #header #header-inner-wrap.no-transparent.top-animate { top: 62px; } body.header-transparent .layout-border #header #header-inner-wrap.no-transparent.top-animate, body.sticky-header .layout-border #header #header-inner-wrap.no-transparent.top-animate { top: 30px; } body.header-transparent.admin-bar .layout-border.layout-border-header-top #header #header-inner-wrap.no-transparent.top-animate, body.sticky-header.admin-bar .layout-border.layout-border-header-top #header #header-inner-wrap.no-transparent.top-animate { top: 32px; z-index: 15; } body.header-transparent .layout-border.layout-border-header-top #header #header-inner-wrap.no-transparent.top-animate, body.sticky-header .layout-border.layout-border-header-top #header #header-inner-wrap.no-transparent.top-animate { top: 0px; z-index: 15; } body.header-transparent .layout-border #header #header-inner-wrap.no-transparent #header-wrap, body.sticky-header .layout-border #header #header-inner-wrap.no-transparent #header-wrap { margin: 0px 30px; -webkit-box-sizing: border-box; -moz-box-sizing: border-box; box-sizing: border-box; position: relative; } .mfp-content.layout-border img { padding: 70px 0px 70px 0px; } body.admin-bar .mfp-content.layout-border img { padding: 102px 0px 70px 0px; } .mfp-content.layout-border .mfp-bottom-bar { margin-top: -60px; } body .mfp-content.layout-border .mfp-close { top: 30px; } body.admin-bar .mfp-content.layout-border .mfp-close { top: 62px; } pre { background-image: -webkit-repeating-linear-gradient(top, #FFFFFF 0px, #FFFFFF 30px, #fafbfd 24px, #fafbfd 56px); background-image: -moz-repeating-linear-gradient(top, #FFFFFF 0px, #FFFFFF 30px, #fafbfd 24px, #fafbfd 56px); background-image: -ms-repeating-linear-gradient(top, #FFFFFF 0px, #FFFFFF 30px, #fafbfd 24px, #fafbfd 56px); background-image: -o-repeating-linear-gradient(top, #FFFFFF 0px, #FFFFFF 30px, #fafbfd 24px, #fafbfd 56px); background-image: repeating-linear-gradient(top, #FFFFFF 0px, #FFFFFF 30px, #fafbfd 24px, #fafbfd 56px); display: block; line-height: 28px; margin-bottom: 50px; overflow: auto; padding: 0px 10px; border:1px solid #eeeeee; } .post-title a{ color: inherit; } /*Animated link Typography*/ .be-sidemenu, .special-header-menu a::before{ background-color: rgb(26,26,26);background-color: rgba(26,26,26,1);} /*For normal styles add the padding in top and bottom*/ .be-themes-layout-layout-border .be-sidemenu, .be-themes-layout-layout-border .be-sidemenu, .be-themes-layout-layout-border-header-top .be-sidemenu, .be-themes-layout-layout-border-header-top .be-sidemenu{ padding: 30px 0px; box-sizing: border-box; } /*For center-align and left-align overlay, add padding to all sides*/ .be-themes-layout-layout-border.overlay-left-align-menu .be-sidemenu, .be-themes-layout-layout-border.overlay-center-align-menu .be-sidemenu, .be-themes-layout-layout-border-header-top.overlay-left-align-menu .be-sidemenu, .be-themes-layout-layout-border-header-top.overlay-center-align-menu .be-sidemenu{ padding: 30px; box-sizing: border-box; } .be-themes-layout-layout-border-header-top .be-sidemenu{ padding-top: 0px; } body.perspective-left.perspectiveview, body.perspective-right.perspectiveview{ background-color: rgb(26,26,26);background-color: rgba(26,26,26,1);} body.left-header.perspective-right.perspectiveview{ background-color: rgb(8,8,8);background-color: rgba(8,8,8,0.90);} body.perspective-left .be-sidemenu, body.perspective-right .be-sidemenu{ background-color : transparent; } /*Portfolio navigation*/ .loader-style1-double-bounce1, .loader-style1-double-bounce2, .loader-style2-wrap, .loader-style3-wrap > div, .loader-style5-wrap .dot1, .loader-style5-wrap .dot2, #nprogress .bar { background: #00bcdd !important; } .loader-style4-wrap { border-top: 7px solid rgba(0, 188, 221 , 0.3); border-right: 7px solid rgba(0, 188, 221 , 0.3); border-bottom: 7px solid rgba(0, 188, 221 , 0.3); border-left-color: #00bcdd; } #nprogress .spinner-icon { border-top-color: #00bcdd !important; border-left-color: #00bcdd !important; } #nprogress .peg { box-shadow: 0 0 10px #00bcdd, 0 0 5px #00bcdd !important; } .style1 #navigation, .style3 #navigation, .style4 #navigation, .style5 #navigation, #header-controls-left, #header-controls-right, #header-wrap, .mobile-nav-controller-wrap, #left-header-mobile .header-cart-controls, .style6 #navigation-left-side, .style6 #navigation-right-side, .style7 #navigation{ line-height: 66px; } body.header-transparent #header-wrap #navigation, body.header-transparent #header-wrap #navigation-left-side, body.header-transparent #header-wrap #navigation-right-side, body.header-transparent #header-inner-wrap .header-controls, body.header-transparent #header-inner-wrap #header-controls-left, body.header-transparent #header-inner-wrap #header-controls-right, body.header-transparent #header-inner-wrap #header-wrap, body.header-transparent #header-inner-wrap .mobile-nav-controller-wrap { line-height: 66px; } body #header-inner-wrap.top-animate #navigation, body #header-inner-wrap.top-animate #navigation-left-side, body #header-inner-wrap.top-animate #navigation-right-side, body #header-inner-wrap.top-animate .header-controls, body #header-inner-wrap.top-animate #header-wrap, body #header-inner-wrap.top-animate #header-controls-right, body #header-inner-wrap.top-animate #header-controls-left { line-height: 66px; } .header-transparent #content.page-split-screen-left, .header-transparent #content.page-split-screen-right{ } #navigation-left-side { padding-right: 90.5px; } #navigation-right-side { padding-left: 90.5px; } #bbpress-forums li.bbp-body ul.forum, #bbpress-forums li.bbp-body ul.topic { border-top: 1px solid #eeeeee; } #bbpress-forums ul.bbp-lead-topic, #bbpress-forums ul.bbp-topics, #bbpress-forums ul.bbp-forums, #bbpress-forums ul.bbp-replies, #bbpress-forums ul.bbp-search-results { border: 1px solid #eeeeee; } #bbpress-forums li.bbp-header, #bbpress-forums li.bbp-footer, .menu-card-item.highlight-menu-item { background: #fafbfd; } #bbpress-forums .topic .bbp-topic-meta a:hover, .bbp-forum-freshness a:hover, .bbp-topic-freshness a:hover, .bbp-header .bbp-reply-content a:hover, .bbp-topic-tags a:hover, .bbp-breadcrumb a:hover, .bbp-forums-list a:hover { color: #00bcdd; } div.bbp-reply-header, .bar-style-related-posts-list, .menu-card-item { border-color: #eeeeee; } #evcal_list .eventon_list_event .evcal_desc span.evcal_event_title, .eventon_events_list .evcal_event_subtitle { padding-bottom: 10px !important; } .eventon_events_list .eventon_list_event .evcal_desc, .evo_pop_body .evcal_desc, #page-content p.evcal_desc { padding-left: 100px !important; } .evcal_evdata_row { background: #fafbfd !important; } .eventon_events_list .eventon_list_event .event_description { background: #fafbfd !important; border-color: #eeeeee !important; } .bordr, #evcal_list .bordb { border-color: #eeeeee !important; } .evcal_evdata_row .evcal_evdata_cell h3 { margin-bottom: 10px !important; } /**** Be single portfolio - overflow images ****/ /* Optiopn Panel Css */ /********************************************** *** FORMS **********************************************/ .mc4wp-form input {height: 50px;} .wpcf7-form textarea { height: 100px; } .wpcf7-submit { height: 50px; } input[type="submit"] { border-radius: 2px; cursor: pointer; } input.wpcf7-form-control.wpcf7-submit.be-button { border: 2px solid; } .hbspt-form label { color: #fff; } .hbspt-form .input { padding-bottom: 15px; } #hsForm_1692b5bd-5639-41c0-9e31-af463bc55211 .actions input[type="submit"], #hsForm_4b36f79d-3c51-4938-bd29-6b08acce3c48 .actions input[type="submit"]{ background-color: #00bcdd; } .actions input[type="submit"] { background-color: #323840; } /********************************************** *** TEAM **********************************************/ .team-description { font-weight: bold; } .speaker-details-circle img { border-radius: 50%; } .speaker-social-icons { margin-bottom: 15px; } /********************************************** *** GENERAL **********************************************/ // .odsc__box-shadow { // box-shadow: 0 2px 4px 0 rgba(0, 0, 0, 0.5); // } .flip-wrap { box-shadow: 0 2px 4px 0 rgba(0, 0, 0, 0.5); } .pricing-table li.pricing-button .be-button { border-radius: 2px; } .color-white { color: #fff; } span.top-bar-widgets { font-size: 105%; font-weight: bold; } .be-nav-link-effect-1 a::after { top: 140%; } .myTable { background-color:#fff;border-collapse:collapse; } .myTable th { background-color:#fff;color:white;width:50%; } .myTable td, .myTable th { padding:5px;border:1px solid #fff; } .menu-card-item-price { line-height: 12px !important; top: 50%; transform: translateY(-50%); width: 40%; } a#back-to-top { border-radius: 4px; border: solid #323840 2px; } @media (min-width: 60rem) { .ai-max-width { max-width: 50rem; margin: 0 auto !important; } } .icon-quote { transform: scaleX(-1) scaleY(-1); } /********************************************** *** MASTERCLASS **********************************************/ .masterclass-hero { color: #fff; text-align: center; max-width: 900px; margin: auto; } @media (min-width: 60rem) { .masterclass-countdown { max-width: 600px; margin: 0 auto !important; } } .masterclass-countdown .countdown-amount { font-size: 45px; } /* Mobile 480 */ @media (max-width: 60rem) { h3.masterclass-hero { font-size: 20px; line-height: 30px; } h4.masterclass-hero { font-size: 16px; } } .white-border { border-right: 3px solid #00bcdd; } .max-width-text { max-width: 100px; margin: 0px auto; } @media (min-width: 60rem) { .max-width-text-main { max-width: 800px; margin: 0px auto !important; } } /* Partnering */ .partnering-white-text { color: #fff; line-height: 1.7; font-weight: bold; font-size: 14pt; } /********************************************** *** ACORDIANS **********************************************/ .accordion .accordion-head { border-bottom: 1px solid; padding-bottom: 1%; } /********************************************** *** BLOG **********************************************/ .display-none { display: none; } .page-id-1424 #header-wrap, .page-id-10 #header-wrap, .page-id-267 #header-wrap, .page-id-1296 #header-wrap, .page-id-1836 #header-wrap, .page-id-218 #header-wrap, .page-id-264 #header-wrap, .page-id-2275 #header-wrap, .home #header-wrap, .search #header-wrap, .post-template-default #header-wrap, .blog #header-wrap, .archive #header-wrap { display: none; } .blog .post-thumb a:hover { border: 2px solid; } /********************************************** *** PORTFILIO **********************************************/ .portfolio .thumb-title-wrap .thumb-title { line-height: 20px; } span.speaker-name { color: #fff; } .speaker-company { text-transform: none; font-size: 75%; color: #323840; } .portfolio hr { margin-bottom: 15px; border-color: #323840; } </style><meta name="onesignal" content="wordpress-plugin"/> <script> window.OneSignal = window.OneSignal || []; OneSignal.push( function() { OneSignal.SERVICE_WORKER_UPDATER_PATH = "OneSignalSDKUpdaterWorker.js.php"; OneSignal.SERVICE_WORKER_PATH = "OneSignalSDKWorker.js.php"; OneSignal.SERVICE_WORKER_PARAM = { scope: "/" }; OneSignal.setDefaultNotificationUrl("https://odsc.com"); var oneSignal_options = {}; window._oneSignalInitOptions = oneSignal_options; oneSignal_options['wordpress'] = true; oneSignal_options['appId'] = 'aef7b140-4d2b-469f-add6-67112f54e19c'; oneSignal_options['allowLocalhostAsSecureOrigin'] = true; oneSignal_options['welcomeNotification'] = { }; oneSignal_options['welcomeNotification']['title'] = ""; oneSignal_options['welcomeNotification']['message'] = ""; oneSignal_options['path'] = "https://odsc.com/wp-content/plugins/onesignal-free-web-push-notifications/sdk_files/"; oneSignal_options['promptOptions'] = { }; oneSignal_options['notifyButton'] = { }; oneSignal_options['notifyButton']['enable'] = true; oneSignal_options['notifyButton']['position'] = 'bottom-right'; oneSignal_options['notifyButton']['theme'] = 'default'; oneSignal_options['notifyButton']['size'] = 'medium'; oneSignal_options['notifyButton']['showCredit'] = true; oneSignal_options['notifyButton']['text'] = {}; OneSignal.init(window._oneSignalInitOptions); OneSignal.showNativePrompt(); }); function documentInitOneSignal() { var oneSignal_elements = document.getElementsByClassName("OneSignal-prompt"); var oneSignalLinkClickHandler = function(event) { OneSignal.push(['registerForPushNotifications']); event.preventDefault(); }; for(var i = 0; i < oneSignal_elements.length; i++) oneSignal_elements[i].addEventListener('click', oneSignalLinkClickHandler, false); } if (document.readyState === 'complete') { documentInitOneSignal(); } else { window.addEventListener("load", function(event){ documentInitOneSignal(); }); } </script> <script>function initApollo(){var n=Math.random().toString(36).substring(7),o=document.createElement("script"); o.src="https://assets.apollo.io/micro/website-tracker/tracker.iife.js?nocache="+n,o.async=!0,o.defer=!0, o.onload=function(){window.trackingFunctions.onLoad({appId:"6631283d8cd2f401c74bd616"})}, document.head.appendChild(o)}initApollo();</script> <script type="text/javascript" src="https://addevent.com/libs/atc/1.6.1/atc.min.js" async defer></script><meta name="google-site-verification" content="mkcmvYzbHoP3pfxm_tVOMEGp0yhOe2G47zBsS01bPGs" /> <script>(function(w,d,t,r,u){var f,n,i;w[u]=w[u]||[],f=function(){var o={ti:"25050152"};o.q=w[u],w[u]=new UET(o),w[u].push("pageLoad")},n=d.createElement(t),n.src=r,n.async=1,n.onload=n.onreadystatechange=function(){var s=this.readyState;s&&s!=="loaded"&&s!=="complete"||(f(),n.onload=n.onreadystatechange=null)},i=d.getElementsByTagName(t)[0],i.parentNode.insertBefore(n,i)})(window,document,"script","//bat.bing.com/bat.js","uetq");</script> <script charset="utf-8" src="https://js.hubspot.com/forms/current.js"></script> <!-- DO NOT MODIFY --> <!-- Quora Pixel Code (JS Helper) --> <script> !function(q,e,v,n,t,s){if(q.qp) return; n=q.qp=function(){n.qp?n.qp.apply(n,arguments):n.queue.push(arguments);}; n.queue=[];t=document.createElement(e);t.async=!0;t.src=v; s=document.getElementsByTagName(e)[0]; s.parentNode.insertBefore(t,s);}(window, 'script', 'https://a.quora.com/qevents.js'); qp('init', '8ad753e5bebf4fb58db2eb33a263c02d'); qp('track', 'ViewContent'); </script> <noscript><img height="1" width="1" style="display:none" src="https://q.quora.com/_/ad/8ad753e5bebf4fb58db2eb33a263c02d/pixel?tag=ViewContent&noscript=1"/></noscript> <!-- End of Quora Pixel Code --><style rel="stylesheet" id="typehub-output">h1,.h1{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:700;font-style:normal;text-transform:none;font-size:55px;line-height:70px;letter-spacing:0;color:#515151}h2,.h2{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:700;font-style:normal;text-transform:none;font-size:42px;line-height:63px;letter-spacing:0;color:#515151}h3,.h3{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:700;font-style:normal;text-transform:none;font-size:35px;line-height:52px;letter-spacing:1px;color:#515151}h4,.woocommerce-order-received .woocommerce h2,.woocommerce-order-received .woocommerce h3,.woocommerce-view-order .woocommerce h2,.woocommerce-view-order .woocommerce h3,.h4{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;text-transform:none;font-size:26px;line-height:42px;letter-spacing:0;color:#515151}h5,#reply-title,.h5{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;text-transform:none;font-size:20px;line-height:36px;letter-spacing:0;color:#515151}h6,.testimonial-author-role.h6-font,.menu-card-title,.menu-card-item-price,.slider-counts,.woocommerce-MyAccount-navigation ul li,a.bbp-forum-title,#bbpress-forums fieldset.bbp-form label,.bbp-topic-title a.bbp-topic-permalink,#bbpress-forums ul.forum-titles li,#bbpress-forums ul.bbp-replies li.bbp-header,.h6{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;text-transform:none;font-size:15px;line-height:32px;letter-spacing:0;color:#515151}body,.special-heading-wrap .caption-wrap .body-font,.woocommerce .woocommerce-ordering select.orderby,.woocommerce-page .woocommerce-ordering select.orderby,.body{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;text-transform:none;font-size:18px;line-height:26px;letter-spacing:0;color:#5f6263}.page-title-module-custom .page-title-custom,h6.portfolio-title-nav{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;text-transform:uppercase;font-size:18px;line-height:36px;letter-spacing:3px;color:#515151}.sub-title,.special-subtitle,.sub_title{font-family:"Crimson Text",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;text-transform:none;font-size:15px;letter-spacing:0}#footer{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;text-transform:none;font-size:14px;line-height:14px;letter-spacing:0;color:#fff}.special-header-menu .menu-container,#navigation .mega .sub-menu .highlight .sf-with-ul,#navigation,.style2 #navigation,.style13 #navigation,#navigation-left-side,#navigation-right-side,.sb-left #slidebar-menu,.header-widgets,.header-code-widgets,body #header-inner-wrap.top-animate.style2 #navigation,.top-overlay-menu .sb-right #slidebar-menu{font-family:"Montserrat",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:600;font-style:normal;text-transform:uppercase;font-size:16px;line-height:51px;letter-spacing:1px;color:#00bcdd}.special-header-menu .menu-container .sub-menu,.special-header-menu .sub-menu,#navigation .sub-menu,#navigation .children,#navigation-left-side .sub-menu,#navigation-left-side .children,#navigation-right-side .sub-menu,#navigation-right-side .children,.sb-left #slidebar-menu .sub-menu,.top-overlay-menu .sb-right #slidebar-menu .sub-menu{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;text-transform:none;font-size:16px;line-height:28px;letter-spacing:0;color:#bbb}ul#mobile-menu a,ul#mobile-menu li.mega ul.sub-menu li.highlight>:first-child{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;text-transform:uppercase;font-size:12px;line-height:40px;letter-spacing:1px;color:#fff}ul#mobile-menu ul.sub-menu a{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;text-transform:none;font-size:13px;line-height:27px;letter-spacing:0;color:#bbb}.top-right-sliding-menu .sb-right ul#slidebar-menu li,.sb-right #slidebar-menu .mega .sub-menu .highlight .sf-with-ul{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;text-transform:uppercase;font-size:16px;line-height:50px;letter-spacing:1px;color:#fff}.top-right-sliding-menu .sb-right #slidebar-menu ul.sub-menu li{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;text-transform:none;font-size:16px;line-height:25px;letter-spacing:0;color:#fff}.ui-tabs-anchor,.accordion .accordion-head,.skill-wrap .skill_name,.chart-wrap span,.animate-number-wrap h6 span,.woocommerce-tabs .tabs li a,.be-countdown{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:600;font-style:normal;letter-spacing:0}.ui-tabs-anchor{text-transform:uppercase;font-size:13px;line-height:17px}.accordion .accordion-head{text-transform:uppercase;font-size:18px;line-height:17px}.skill-wrap .skill_name{text-transform:none;font-size:20px;line-height:17px}.countdown-amount{text-transform:uppercase;font-size:55px;line-height:95px}.countdown-section{text-transform:uppercase;font-size:15px;line-height:30px}.testimonial_slide .testimonial-content{font-family:"Crimson Text",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;text-transform:none;letter-spacing:0}.tweet-slides .tweet-content{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;text-transform:none;letter-spacing:0}.tatsu-button,.be-button,.woocommerce a.button,.woocommerce-page a.button,.woocommerce button.button,.woocommerce-page button.button,.woocommerce input.button,.woocommerce-page input.button,.woocommerce #respond input#submit,.woocommerce-page #respond input#submit,.woocommerce #content input.button,.woocommerce-page #content input.button,input[type="submit"],.more-link.style1-button,.more-link.style2-button,.more-link.style3-button,input[type="button"],input[type="submit"],input[type="reset"],button,input[type="file"]::-webkit-file-upload-button{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif}.oshine-animated-link,.view-project-link.style4-button{font-family:"Montserrat",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;text-transform:none;letter-spacing:0}.thumb-title-wrap .thumb-title,.full-screen-portfolio-overlay-title{font-family:"Montserrat",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;font-size:14px;line-height:30px}.thumb-title-wrap .portfolio-item-cats{font-size:12px;line-height:17px}h6.gallery-side-heading{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;text-transform:none;font-size:15px;line-height:32px;letter-spacing:0;color:#515151}.portfolio-details .gallery-side-heading-wrap p{font-family:"Raleway",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;font-size:13px;line-height:26px}a.navigation-previous-post-link,a.navigation-next-post-link{font-family:"Montserrat",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:700;font-style:normal;font-size:13px;line-height:20px}#portfolio-title-nav-bottom-wrap h6,#portfolio-title-nav-bottom-wrap .slider-counts{font-family:"Montserrat",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;font-size:15px}.attachment-details-custom-slider{font-family:"Crimson Text",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-size:15px;line-height:15px;color:}.filters .filter_item{font-family:"Montserrat",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;font-size:12px;line-height:32px;color:#222}.woocommerce ul.products li.product .product-meta-data h3,.woocommerce-page ul.products li.product .product-meta-data h3,.woocommerce ul.products li.product h3,.woocommerce-page ul.products li.product h3,.woocommerce ul.products li.product .product-meta-data .woocommerce-loop-product__title,.woocommerce-page ul.products li.product .product-meta-data .woocommerce-loop-product__title,.woocommerce ul.products li.product .woocommerce-loop-product__title,.woocommerce-page ul.products li.product .woocommerce-loop-product__title,.woocommerce ul.products li.product-category .woocommerce-loop-category__title,.woocommerce-page ul.products li.product-category .woocommerce-loop-category__title{font-family:"Montserrat",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;text-transform:uppercase;font-size:13px;line-height:27px;letter-spacing:1px;color:#222}.woocommerce-page.single.single-product #content div.product h1.product_title.entry-title{font-family:"Montserrat",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;text-transform:none;font-size:25px;line-height:27px;letter-spacing:0;color:#222}.post-title,.post-date-wrap{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;text-transform:none;font-size:20px;line-height:40px;letter-spacing:0;color:#000}.style3-blog .post-title,.style8-blog .post-title{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;text-transform:none;font-size:16px;line-height:28px;letter-spacing:0;color:#363c3b}.post-meta.post-top-meta-typo,.style8-blog .post-meta.post-category a,.hero-section-blog-categories-wrap a{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;text-transform:uppercase;font-size:12px;line-height:24px;letter-spacing:0;color:#757575}.post-nav li,.style8-blog .post-meta.post-date,.style8-blog .post-bottom-meta-wrap,.hero-section-blog-bottom-meta-wrap{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;text-transform:uppercase;font-size:12px;line-height:24px;letter-spacing:0;color:#757575}.single-post .post-title,.single-post .style3-blog .post-title,.single-post .style8-blog .post-title{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;text-transform:none;font-size:20px;line-height:40px;letter-spacing:0;color:#000}.sidebar-widgets h6{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;text-transform:uppercase;font-size:18px;line-height:22px;letter-spacing:1px;color:#333}.sidebar-widgets{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;text-transform:none;font-size:16px;line-height:24px;letter-spacing:0;color:#606060}.sb-slidebar .widget h6{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;text-transform:none;font-size:16px;line-height:22px;letter-spacing:1px;color:#fff}.sb-slidebar .widget{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;text-transform:none;font-size:14px;line-height:25px;letter-spacing:0;color:#a2a2a2}#bottom-widgets h6{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;text-transform:uppercase;font-size:18px;line-height:22px;letter-spacing:1px;color:#474747}#bottom-widgets{font-family:"Source Sans Pro",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;text-transform:none;font-size:16px;line-height:24px;letter-spacing:0;color:#757575}.contact_form_module input[type="text"],.contact_form_module textarea{font-family:"Montserrat",-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,Oxygen-Sans,Ubuntu,Cantarell,'Helvetica Neue',sans-serif;font-weight:400;font-style:normal;text-transform:none;font-size:13px;line-height:26px;letter-spacing:0;color:#222}@media only screen and (max-width:1377px){}@media only screen and (min-width:768px) and (max-width:1024px){}@media only screen and (max-width:767px){}</style><style id = "oshine-typehub-css" type="text/css"> #header-controls-right,#header-controls-left{color:#00bcdd}#be-left-strip .be-mobile-menu-icon span{background-color:#323232}ul#mobile-menu .mobile-sub-menu-controller{line-height :40px }ul#mobile-menu ul.sub-menu .mobile-sub-menu-controller{line-height :27px }.breadcrumbs{color:#515151}.search-box-wrapper.style2-header-search-widget input[type="text"]{font-style:normal;font-weight:400;font-family:Crimson Text}.portfolio-share a.custom-share-button,.portfolio-share a.custom-share-button:active,.portfolio-share a.custom-share-button:hover,.portfolio-share a.custom-share-button:visited{color:#515151}.more-link.style2-button{color:#000 !important;border-color:#000 !important}.style8-blog .post-bottom-meta-wrap .be-share-stack a.custom-share-button,.style8-blog .post-bottom-meta-wrap .be-share-stack a.custom-share-button:active,.style8-blog .post-bottom-meta-wrap .be-share-stack a.custom-share-button:hover,.style8-blog .post-bottom-meta-wrap .be-share-stack a.custom-share-button:visited{color:#757575}.hero-section-blog-categories-wrap a,.hero-section-blog-categories-wrap a:visited,.hero-section-blog-categories-wrap a:hover,.hero-section-blog-bottom-meta-wrap .hero-section-blog-bottom-meta-wrap a,.hero-section-blog-bottom-meta-wrap a:visited,.hero-section-blog-bottom-meta-wrap a:hover,.hero-section-blog-bottom-meta-wrap{color :#000}#navigation .mega .sub-menu .highlight .sf-with-ul{color:#bbb !important;line-height:1.5}.view-project-link.style4-button{color :#515151}.related.products h2,.upsells.products h2,.cart-collaterals .cross-sells h2,.cart_totals h2,.shipping_calculator h2,.woocommerce-billing-fields h3,.woocommerce-shipping-fields h3,.shipping_calculator h2,#order_review_heading,.woocommerce .page-title{font-family:Montserrat;font-weight:400}.woocommerce form .form-row label,.woocommerce-page form .form-row label{color:#515151}.woocommerce-tabs .tabs li a{color:#515151 !important}#bbpress-forums ul.forum-titles li,#bbpress-forums ul.bbp-replies li.bbp-header{line-height:inherit;letter-spacing:inherit;text-transform:uppercase;font-size:inherit}#bbpress-forums .topic .bbp-topic-meta a,.bbp-forum-freshness a,.bbp-topic-freshness a,.bbp-header .bbp-reply-content a,.bbp-topic-tags a,.bbp-breadcrumb a,.bbp-forums-list a{color:#515151}.ajde_evcal_calendar .calendar_header p,.eventon_events_list .eventon_list_event .evcal_cblock{font-family:Source Sans Pro !important}.eventon_events_list .eventon_list_event .evcal_desc span.evcal_desc2,.evo_pop_body .evcal_desc span.evcal_desc2{font-family:Source Sans Pro !important;font-size:14px !important;text-transform:none}.eventon_events_list .eventon_list_event .evcal_desc span.evcal_event_subtitle,.evo_pop_body .evcal_desc span.evcal_event_subtitle,.evcal_evdata_row .evcal_evdata_cell p,#evcal_list .eventon_list_event p.no_events{text-transform:none !important;font-family:Source Sans Pro !important;font-size:inherit !important} </style><style rel="stylesheet" id="colorhub-output">.swatch-red, .swatch-red a{color:#f44336;}.swatch-pink, .swatch-pink a{color:#E91E63;}.swatch-purple, .swatch-purple a{color:#9C27B0;}.swatch-deep-purple, .swatch-deep-purple a{color:#673AB7;}.swatch-indigo, .swatch-indigo a{color:#3F51B5;}.swatch-blue, .swatch-blue a{color:#2196F3;}.swatch-light-blue, .swatch-light-blue a{color:#03A9F4;}.swatch-cyan, .swatch-cyan a{color:#00BCD4;}.swatch-teal, .swatch-teal a{color:#009688;}.swatch-green, .swatch-green a{color:#4CAF50;}.swatch-light-green, .swatch-light-green a{color:#8BC34A;}.swatch-lime, .swatch-lime a{color:#CDDC39;}.swatch-yellow, .swatch-yellow a{color:#FFEB3B;}.swatch-amber, .swatch-amber a{color:#FFC107;}.swatch-orange, .swatch-orange a{color:#FF9800;}.swatch-deep-orange, .swatch-deep-orange a{color:#FF5722;}.swatch-brown, .swatch-brown a{color:#795548;}.swatch-grey, .swatch-grey a{color:#9E9E9E;}.swatch-blue-grey, .swatch-blue-grey a{color:#607D8B;}.swatch-white, .swatch-white a{color:#ffffff;}.swatch-black, .swatch-black a{color:#000000;}.palette-0, .palette-0 a{color:#00bcdd;}.palette-1, .palette-1 a{color:#ffffff;}.palette-2, .palette-2 a{color:#222222;}.palette-3, .palette-3 a{color:#888888;}.palette-4, .palette-4 a{color:#fafbfd;}</style><link rel="icon" href="https://odsc.com/wp-content/uploads/2022/03/cropped-faviODSC-32x32.png" sizes="32x32" /><link rel="icon" href="https://odsc.com/wp-content/uploads/2022/03/cropped-faviODSC-192x192.png" sizes="192x192" /><link rel="apple-touch-icon" href="https://odsc.com/wp-content/uploads/2022/03/cropped-faviODSC-180x180.png" /><meta name="msapplication-TileImage" content="https://odsc.com/wp-content/uploads/2022/03/cropped-faviODSC-270x270.png" /> <!-- Fonts Plugin CSS - https://fontsplugin.com/ --> <style> /* Cached: November 22, 2024 at 8:21am */ /* cyrillic-ext */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 100; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOiCnqEu92Fr1Mu51QrEz0dL_nz.woff2) format('woff2'); unicode-range: U+0460-052F, U+1C80-1C8A, U+20B4, U+2DE0-2DFF, U+A640-A69F, U+FE2E-FE2F; } /* cyrillic */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 100; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOiCnqEu92Fr1Mu51QrEzQdL_nz.woff2) format('woff2'); unicode-range: U+0301, U+0400-045F, U+0490-0491, U+04B0-04B1, U+2116; } /* greek-ext */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 100; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOiCnqEu92Fr1Mu51QrEzwdL_nz.woff2) format('woff2'); unicode-range: U+1F00-1FFF; } /* greek */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 100; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOiCnqEu92Fr1Mu51QrEzMdL_nz.woff2) format('woff2'); unicode-range: U+0370-0377, U+037A-037F, U+0384-038A, U+038C, U+038E-03A1, U+03A3-03FF; } /* vietnamese */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 100; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOiCnqEu92Fr1Mu51QrEz8dL_nz.woff2) format('woff2'); unicode-range: U+0102-0103, U+0110-0111, U+0128-0129, U+0168-0169, U+01A0-01A1, U+01AF-01B0, U+0300-0301, U+0303-0304, U+0308-0309, U+0323, U+0329, U+1EA0-1EF9, U+20AB; } /* latin-ext */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 100; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOiCnqEu92Fr1Mu51QrEz4dL_nz.woff2) format('woff2'); unicode-range: U+0100-02BA, U+02BD-02C5, U+02C7-02CC, U+02CE-02D7, U+02DD-02FF, U+0304, U+0308, U+0329, U+1D00-1DBF, U+1E00-1E9F, U+1EF2-1EFF, U+2020, U+20A0-20AB, U+20AD-20C0, U+2113, U+2C60-2C7F, U+A720-A7FF; } /* latin */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 100; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOiCnqEu92Fr1Mu51QrEzAdLw.woff2) format('woff2'); unicode-range: U+0000-00FF, U+0131, U+0152-0153, U+02BB-02BC, U+02C6, U+02DA, U+02DC, U+0304, U+0308, U+0329, U+2000-206F, U+20AC, U+2122, U+2191, U+2193, U+2212, U+2215, U+FEFF, U+FFFD; } /* cyrillic-ext */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 300; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51TjASc3CsTKlA.woff2) format('woff2'); unicode-range: U+0460-052F, U+1C80-1C8A, U+20B4, U+2DE0-2DFF, U+A640-A69F, U+FE2E-FE2F; } /* cyrillic */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 300; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51TjASc-CsTKlA.woff2) format('woff2'); unicode-range: U+0301, U+0400-045F, U+0490-0491, U+04B0-04B1, U+2116; } /* greek-ext */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 300; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51TjASc2CsTKlA.woff2) format('woff2'); unicode-range: U+1F00-1FFF; } /* greek */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 300; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51TjASc5CsTKlA.woff2) format('woff2'); unicode-range: U+0370-0377, U+037A-037F, U+0384-038A, U+038C, U+038E-03A1, U+03A3-03FF; } /* vietnamese */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 300; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51TjASc1CsTKlA.woff2) format('woff2'); unicode-range: U+0102-0103, U+0110-0111, U+0128-0129, U+0168-0169, U+01A0-01A1, U+01AF-01B0, U+0300-0301, U+0303-0304, U+0308-0309, U+0323, U+0329, U+1EA0-1EF9, U+20AB; } /* latin-ext */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 300; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51TjASc0CsTKlA.woff2) format('woff2'); unicode-range: U+0100-02BA, U+02BD-02C5, U+02C7-02CC, U+02CE-02D7, U+02DD-02FF, U+0304, U+0308, U+0329, U+1D00-1DBF, U+1E00-1E9F, U+1EF2-1EFF, U+2020, U+20A0-20AB, U+20AD-20C0, U+2113, U+2C60-2C7F, U+A720-A7FF; } /* latin */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 300; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51TjASc6CsQ.woff2) format('woff2'); unicode-range: U+0000-00FF, U+0131, U+0152-0153, U+02BB-02BC, U+02C6, U+02DA, U+02DC, U+0304, U+0308, U+0329, U+2000-206F, U+20AC, U+2122, U+2191, U+2193, U+2212, U+2215, U+FEFF, U+FFFD; } /* cyrillic-ext */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 400; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOkCnqEu92Fr1Mu51xFIzIFKw.woff2) format('woff2'); unicode-range: U+0460-052F, U+1C80-1C8A, U+20B4, U+2DE0-2DFF, U+A640-A69F, U+FE2E-FE2F; } /* cyrillic */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 400; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOkCnqEu92Fr1Mu51xMIzIFKw.woff2) format('woff2'); unicode-range: U+0301, U+0400-045F, U+0490-0491, U+04B0-04B1, U+2116; } /* greek-ext */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 400; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOkCnqEu92Fr1Mu51xEIzIFKw.woff2) format('woff2'); unicode-range: U+1F00-1FFF; } /* greek */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 400; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOkCnqEu92Fr1Mu51xLIzIFKw.woff2) format('woff2'); unicode-range: U+0370-0377, U+037A-037F, U+0384-038A, U+038C, U+038E-03A1, U+03A3-03FF; } /* vietnamese */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 400; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOkCnqEu92Fr1Mu51xHIzIFKw.woff2) format('woff2'); unicode-range: U+0102-0103, U+0110-0111, U+0128-0129, U+0168-0169, U+01A0-01A1, U+01AF-01B0, U+0300-0301, U+0303-0304, U+0308-0309, U+0323, U+0329, U+1EA0-1EF9, U+20AB; } /* latin-ext */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 400; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOkCnqEu92Fr1Mu51xGIzIFKw.woff2) format('woff2'); unicode-range: U+0100-02BA, U+02BD-02C5, U+02C7-02CC, U+02CE-02D7, U+02DD-02FF, U+0304, U+0308, U+0329, U+1D00-1DBF, U+1E00-1E9F, U+1EF2-1EFF, U+2020, U+20A0-20AB, U+20AD-20C0, U+2113, U+2C60-2C7F, U+A720-A7FF; } /* latin */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 400; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOkCnqEu92Fr1Mu51xIIzI.woff2) format('woff2'); unicode-range: U+0000-00FF, U+0131, U+0152-0153, U+02BB-02BC, U+02C6, U+02DA, U+02DC, U+0304, U+0308, U+0329, U+2000-206F, U+20AC, U+2122, U+2191, U+2193, U+2212, U+2215, U+FEFF, U+FFFD; } /* cyrillic-ext */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 500; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51S7ACc3CsTKlA.woff2) format('woff2'); unicode-range: U+0460-052F, U+1C80-1C8A, U+20B4, U+2DE0-2DFF, U+A640-A69F, U+FE2E-FE2F; } /* cyrillic */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 500; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51S7ACc-CsTKlA.woff2) format('woff2'); unicode-range: U+0301, U+0400-045F, U+0490-0491, U+04B0-04B1, U+2116; } /* greek-ext */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 500; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51S7ACc2CsTKlA.woff2) format('woff2'); unicode-range: U+1F00-1FFF; } /* greek */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 500; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51S7ACc5CsTKlA.woff2) format('woff2'); unicode-range: U+0370-0377, U+037A-037F, U+0384-038A, U+038C, U+038E-03A1, U+03A3-03FF; } /* vietnamese */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 500; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51S7ACc1CsTKlA.woff2) format('woff2'); unicode-range: U+0102-0103, U+0110-0111, U+0128-0129, U+0168-0169, U+01A0-01A1, U+01AF-01B0, U+0300-0301, U+0303-0304, U+0308-0309, U+0323, U+0329, U+1EA0-1EF9, U+20AB; } /* latin-ext */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 500; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51S7ACc0CsTKlA.woff2) format('woff2'); unicode-range: U+0100-02BA, U+02BD-02C5, U+02C7-02CC, U+02CE-02D7, U+02DD-02FF, U+0304, U+0308, U+0329, U+1D00-1DBF, U+1E00-1E9F, U+1EF2-1EFF, U+2020, U+20A0-20AB, U+20AD-20C0, U+2113, U+2C60-2C7F, U+A720-A7FF; } /* latin */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 500; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51S7ACc6CsQ.woff2) format('woff2'); unicode-range: U+0000-00FF, U+0131, U+0152-0153, U+02BB-02BC, U+02C6, U+02DA, U+02DC, U+0304, U+0308, U+0329, U+2000-206F, U+20AC, U+2122, U+2191, U+2193, U+2212, U+2215, U+FEFF, U+FFFD; } /* cyrillic-ext */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 700; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51TzBic3CsTKlA.woff2) format('woff2'); unicode-range: U+0460-052F, U+1C80-1C8A, U+20B4, U+2DE0-2DFF, U+A640-A69F, U+FE2E-FE2F; } /* cyrillic */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 700; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51TzBic-CsTKlA.woff2) format('woff2'); unicode-range: U+0301, U+0400-045F, U+0490-0491, U+04B0-04B1, U+2116; } /* greek-ext */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 700; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51TzBic2CsTKlA.woff2) format('woff2'); unicode-range: U+1F00-1FFF; } /* greek */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 700; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51TzBic5CsTKlA.woff2) format('woff2'); unicode-range: U+0370-0377, U+037A-037F, U+0384-038A, U+038C, U+038E-03A1, U+03A3-03FF; } /* vietnamese */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 700; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51TzBic1CsTKlA.woff2) format('woff2'); unicode-range: U+0102-0103, U+0110-0111, U+0128-0129, U+0168-0169, U+01A0-01A1, U+01AF-01B0, U+0300-0301, U+0303-0304, U+0308-0309, U+0323, U+0329, U+1EA0-1EF9, U+20AB; } /* latin-ext */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 700; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51TzBic0CsTKlA.woff2) format('woff2'); unicode-range: U+0100-02BA, U+02BD-02C5, U+02C7-02CC, U+02CE-02D7, U+02DD-02FF, U+0304, U+0308, U+0329, U+1D00-1DBF, U+1E00-1E9F, U+1EF2-1EFF, U+2020, U+20A0-20AB, U+20AD-20C0, U+2113, U+2C60-2C7F, U+A720-A7FF; } /* latin */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 700; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51TzBic6CsQ.woff2) format('woff2'); unicode-range: U+0000-00FF, U+0131, U+0152-0153, U+02BB-02BC, U+02C6, U+02DA, U+02DC, U+0304, U+0308, U+0329, U+2000-206F, U+20AC, U+2122, U+2191, U+2193, U+2212, U+2215, U+FEFF, U+FFFD; } /* cyrillic-ext */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 900; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51TLBCc3CsTKlA.woff2) format('woff2'); unicode-range: U+0460-052F, U+1C80-1C8A, U+20B4, U+2DE0-2DFF, U+A640-A69F, U+FE2E-FE2F; } /* cyrillic */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 900; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51TLBCc-CsTKlA.woff2) format('woff2'); unicode-range: U+0301, U+0400-045F, U+0490-0491, U+04B0-04B1, U+2116; } /* greek-ext */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 900; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51TLBCc2CsTKlA.woff2) format('woff2'); unicode-range: U+1F00-1FFF; } /* greek */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 900; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51TLBCc5CsTKlA.woff2) format('woff2'); unicode-range: U+0370-0377, U+037A-037F, U+0384-038A, U+038C, U+038E-03A1, U+03A3-03FF; } /* vietnamese */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 900; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51TLBCc1CsTKlA.woff2) format('woff2'); unicode-range: U+0102-0103, U+0110-0111, U+0128-0129, U+0168-0169, U+01A0-01A1, U+01AF-01B0, U+0300-0301, U+0303-0304, U+0308-0309, U+0323, U+0329, U+1EA0-1EF9, U+20AB; } /* latin-ext */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 900; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51TLBCc0CsTKlA.woff2) format('woff2'); unicode-range: U+0100-02BA, U+02BD-02C5, U+02C7-02CC, U+02CE-02D7, U+02DD-02FF, U+0304, U+0308, U+0329, U+1D00-1DBF, U+1E00-1E9F, U+1EF2-1EFF, U+2020, U+20A0-20AB, U+20AD-20C0, U+2113, U+2C60-2C7F, U+A720-A7FF; } /* latin */ @font-face { font-family: 'Roboto'; font-style: italic; font-weight: 900; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOjCnqEu92Fr1Mu51TLBCc6CsQ.woff2) format('woff2'); unicode-range: U+0000-00FF, U+0131, U+0152-0153, U+02BB-02BC, U+02C6, U+02DA, U+02DC, U+0304, U+0308, U+0329, U+2000-206F, U+20AC, U+2122, U+2191, U+2193, U+2212, U+2215, U+FEFF, U+FFFD; } /* cyrillic-ext */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 100; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOkCnqEu92Fr1MmgVxFIzIFKw.woff2) format('woff2'); unicode-range: U+0460-052F, U+1C80-1C8A, U+20B4, U+2DE0-2DFF, U+A640-A69F, U+FE2E-FE2F; } /* cyrillic */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 100; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOkCnqEu92Fr1MmgVxMIzIFKw.woff2) format('woff2'); unicode-range: U+0301, U+0400-045F, U+0490-0491, U+04B0-04B1, U+2116; } /* greek-ext */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 100; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOkCnqEu92Fr1MmgVxEIzIFKw.woff2) format('woff2'); unicode-range: U+1F00-1FFF; } /* greek */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 100; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOkCnqEu92Fr1MmgVxLIzIFKw.woff2) format('woff2'); unicode-range: U+0370-0377, U+037A-037F, U+0384-038A, U+038C, U+038E-03A1, U+03A3-03FF; } /* vietnamese */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 100; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOkCnqEu92Fr1MmgVxHIzIFKw.woff2) format('woff2'); unicode-range: U+0102-0103, U+0110-0111, U+0128-0129, U+0168-0169, U+01A0-01A1, U+01AF-01B0, U+0300-0301, U+0303-0304, U+0308-0309, U+0323, U+0329, U+1EA0-1EF9, U+20AB; } /* latin-ext */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 100; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOkCnqEu92Fr1MmgVxGIzIFKw.woff2) format('woff2'); unicode-range: U+0100-02BA, U+02BD-02C5, U+02C7-02CC, U+02CE-02D7, U+02DD-02FF, U+0304, U+0308, U+0329, U+1D00-1DBF, U+1E00-1E9F, U+1EF2-1EFF, U+2020, U+20A0-20AB, U+20AD-20C0, U+2113, U+2C60-2C7F, U+A720-A7FF; } /* latin */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 100; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOkCnqEu92Fr1MmgVxIIzI.woff2) format('woff2'); unicode-range: U+0000-00FF, U+0131, U+0152-0153, U+02BB-02BC, U+02C6, U+02DA, U+02DC, U+0304, U+0308, U+0329, U+2000-206F, U+20AC, U+2122, U+2191, U+2193, U+2212, U+2215, U+FEFF, U+FFFD; } /* cyrillic-ext */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 300; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmSU5fCRc4EsA.woff2) format('woff2'); unicode-range: U+0460-052F, U+1C80-1C8A, U+20B4, U+2DE0-2DFF, U+A640-A69F, U+FE2E-FE2F; } /* cyrillic */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 300; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmSU5fABc4EsA.woff2) format('woff2'); unicode-range: U+0301, U+0400-045F, U+0490-0491, U+04B0-04B1, U+2116; } /* greek-ext */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 300; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmSU5fCBc4EsA.woff2) format('woff2'); unicode-range: U+1F00-1FFF; } /* greek */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 300; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmSU5fBxc4EsA.woff2) format('woff2'); unicode-range: U+0370-0377, U+037A-037F, U+0384-038A, U+038C, U+038E-03A1, U+03A3-03FF; } /* vietnamese */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 300; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmSU5fCxc4EsA.woff2) format('woff2'); unicode-range: U+0102-0103, U+0110-0111, U+0128-0129, U+0168-0169, U+01A0-01A1, U+01AF-01B0, U+0300-0301, U+0303-0304, U+0308-0309, U+0323, U+0329, U+1EA0-1EF9, U+20AB; } /* latin-ext */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 300; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmSU5fChc4EsA.woff2) format('woff2'); unicode-range: U+0100-02BA, U+02BD-02C5, U+02C7-02CC, U+02CE-02D7, U+02DD-02FF, U+0304, U+0308, U+0329, U+1D00-1DBF, U+1E00-1E9F, U+1EF2-1EFF, U+2020, U+20A0-20AB, U+20AD-20C0, U+2113, U+2C60-2C7F, U+A720-A7FF; } /* latin */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 300; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmSU5fBBc4.woff2) format('woff2'); unicode-range: U+0000-00FF, U+0131, U+0152-0153, U+02BB-02BC, U+02C6, U+02DA, U+02DC, U+0304, U+0308, U+0329, U+2000-206F, U+20AC, U+2122, U+2191, U+2193, U+2212, U+2215, U+FEFF, U+FFFD; } /* cyrillic-ext */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 400; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOmCnqEu92Fr1Mu72xKOzY.woff2) format('woff2'); unicode-range: U+0460-052F, U+1C80-1C8A, U+20B4, U+2DE0-2DFF, U+A640-A69F, U+FE2E-FE2F; } /* cyrillic */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 400; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOmCnqEu92Fr1Mu5mxKOzY.woff2) format('woff2'); unicode-range: U+0301, U+0400-045F, U+0490-0491, U+04B0-04B1, U+2116; } /* greek-ext */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 400; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOmCnqEu92Fr1Mu7mxKOzY.woff2) format('woff2'); unicode-range: U+1F00-1FFF; } /* greek */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 400; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOmCnqEu92Fr1Mu4WxKOzY.woff2) format('woff2'); unicode-range: U+0370-0377, U+037A-037F, U+0384-038A, U+038C, U+038E-03A1, U+03A3-03FF; } /* vietnamese */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 400; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOmCnqEu92Fr1Mu7WxKOzY.woff2) format('woff2'); unicode-range: U+0102-0103, U+0110-0111, U+0128-0129, U+0168-0169, U+01A0-01A1, U+01AF-01B0, U+0300-0301, U+0303-0304, U+0308-0309, U+0323, U+0329, U+1EA0-1EF9, U+20AB; } /* latin-ext */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 400; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOmCnqEu92Fr1Mu7GxKOzY.woff2) format('woff2'); unicode-range: U+0100-02BA, U+02BD-02C5, U+02C7-02CC, U+02CE-02D7, U+02DD-02FF, U+0304, U+0308, U+0329, U+1D00-1DBF, U+1E00-1E9F, U+1EF2-1EFF, U+2020, U+20A0-20AB, U+20AD-20C0, U+2113, U+2C60-2C7F, U+A720-A7FF; } /* latin */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 400; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOmCnqEu92Fr1Mu4mxK.woff2) format('woff2'); unicode-range: U+0000-00FF, U+0131, U+0152-0153, U+02BB-02BC, U+02C6, U+02DA, U+02DC, U+0304, U+0308, U+0329, U+2000-206F, U+20AC, U+2122, U+2191, U+2193, U+2212, U+2215, U+FEFF, U+FFFD; } /* cyrillic-ext */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 500; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmEU9fCRc4EsA.woff2) format('woff2'); unicode-range: U+0460-052F, U+1C80-1C8A, U+20B4, U+2DE0-2DFF, U+A640-A69F, U+FE2E-FE2F; } /* cyrillic */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 500; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmEU9fABc4EsA.woff2) format('woff2'); unicode-range: U+0301, U+0400-045F, U+0490-0491, U+04B0-04B1, U+2116; } /* greek-ext */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 500; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmEU9fCBc4EsA.woff2) format('woff2'); unicode-range: U+1F00-1FFF; } /* greek */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 500; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmEU9fBxc4EsA.woff2) format('woff2'); unicode-range: U+0370-0377, U+037A-037F, U+0384-038A, U+038C, U+038E-03A1, U+03A3-03FF; } /* vietnamese */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 500; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmEU9fCxc4EsA.woff2) format('woff2'); unicode-range: U+0102-0103, U+0110-0111, U+0128-0129, U+0168-0169, U+01A0-01A1, U+01AF-01B0, U+0300-0301, U+0303-0304, U+0308-0309, U+0323, U+0329, U+1EA0-1EF9, U+20AB; } /* latin-ext */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 500; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmEU9fChc4EsA.woff2) format('woff2'); unicode-range: U+0100-02BA, U+02BD-02C5, U+02C7-02CC, U+02CE-02D7, U+02DD-02FF, U+0304, U+0308, U+0329, U+1D00-1DBF, U+1E00-1E9F, U+1EF2-1EFF, U+2020, U+20A0-20AB, U+20AD-20C0, U+2113, U+2C60-2C7F, U+A720-A7FF; } /* latin */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 500; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmEU9fBBc4.woff2) format('woff2'); unicode-range: U+0000-00FF, U+0131, U+0152-0153, U+02BB-02BC, U+02C6, U+02DA, U+02DC, U+0304, U+0308, U+0329, U+2000-206F, U+20AC, U+2122, U+2191, U+2193, U+2212, U+2215, U+FEFF, U+FFFD; } /* cyrillic-ext */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 700; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmWUlfCRc4EsA.woff2) format('woff2'); unicode-range: U+0460-052F, U+1C80-1C8A, U+20B4, U+2DE0-2DFF, U+A640-A69F, U+FE2E-FE2F; } /* cyrillic */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 700; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmWUlfABc4EsA.woff2) format('woff2'); unicode-range: U+0301, U+0400-045F, U+0490-0491, U+04B0-04B1, U+2116; } /* greek-ext */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 700; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmWUlfCBc4EsA.woff2) format('woff2'); unicode-range: U+1F00-1FFF; } /* greek */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 700; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmWUlfBxc4EsA.woff2) format('woff2'); unicode-range: U+0370-0377, U+037A-037F, U+0384-038A, U+038C, U+038E-03A1, U+03A3-03FF; } /* vietnamese */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 700; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmWUlfCxc4EsA.woff2) format('woff2'); unicode-range: U+0102-0103, U+0110-0111, U+0128-0129, U+0168-0169, U+01A0-01A1, U+01AF-01B0, U+0300-0301, U+0303-0304, U+0308-0309, U+0323, U+0329, U+1EA0-1EF9, U+20AB; } /* latin-ext */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 700; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmWUlfChc4EsA.woff2) format('woff2'); unicode-range: U+0100-02BA, U+02BD-02C5, U+02C7-02CC, U+02CE-02D7, U+02DD-02FF, U+0304, U+0308, U+0329, U+1D00-1DBF, U+1E00-1E9F, U+1EF2-1EFF, U+2020, U+20A0-20AB, U+20AD-20C0, U+2113, U+2C60-2C7F, U+A720-A7FF; } /* latin */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 700; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmWUlfBBc4.woff2) format('woff2'); unicode-range: U+0000-00FF, U+0131, U+0152-0153, U+02BB-02BC, U+02C6, U+02DA, U+02DC, U+0304, U+0308, U+0329, U+2000-206F, U+20AC, U+2122, U+2191, U+2193, U+2212, U+2215, U+FEFF, U+FFFD; } /* cyrillic-ext */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 900; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmYUtfCRc4EsA.woff2) format('woff2'); unicode-range: U+0460-052F, U+1C80-1C8A, U+20B4, U+2DE0-2DFF, U+A640-A69F, U+FE2E-FE2F; } /* cyrillic */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 900; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmYUtfABc4EsA.woff2) format('woff2'); unicode-range: U+0301, U+0400-045F, U+0490-0491, U+04B0-04B1, U+2116; } /* greek-ext */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 900; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmYUtfCBc4EsA.woff2) format('woff2'); unicode-range: U+1F00-1FFF; } /* greek */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 900; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmYUtfBxc4EsA.woff2) format('woff2'); unicode-range: U+0370-0377, U+037A-037F, U+0384-038A, U+038C, U+038E-03A1, U+03A3-03FF; } /* vietnamese */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 900; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmYUtfCxc4EsA.woff2) format('woff2'); unicode-range: U+0102-0103, U+0110-0111, U+0128-0129, U+0168-0169, U+01A0-01A1, U+01AF-01B0, U+0300-0301, U+0303-0304, U+0308-0309, U+0323, U+0329, U+1EA0-1EF9, U+20AB; } /* latin-ext */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 900; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmYUtfChc4EsA.woff2) format('woff2'); unicode-range: U+0100-02BA, U+02BD-02C5, U+02C7-02CC, U+02CE-02D7, U+02DD-02FF, U+0304, U+0308, U+0329, U+1D00-1DBF, U+1E00-1E9F, U+1EF2-1EFF, U+2020, U+20A0-20AB, U+20AD-20C0, U+2113, U+2C60-2C7F, U+A720-A7FF; } /* latin */ @font-face { font-family: 'Roboto'; font-style: normal; font-weight: 900; font-display: swap; src: url(https://fonts.gstatic.com/s/roboto/v32/KFOlCnqEu92Fr1MmYUtfBBc4.woff2) format('woff2'); unicode-range: U+0000-00FF, U+0131, U+0152-0153, U+02BB-02BC, U+02C6, U+02DA, U+02DC, U+0304, U+0308, U+0329, U+2000-206F, U+20AC, U+2122, U+2191, U+2193, U+2212, U+2215, U+FEFF, U+FFFD; } :root { --font-base: Roboto; --font-headings: Roboto; --font-input: Roboto; } body, #content, .entry-content, .post-content, .page-content, .post-excerpt, .entry-summary, .entry-excerpt, .widget-area, .widget, .sidebar, #sidebar, footer, .footer, #footer, .site-footer { font-family: "Roboto"; } #site-title, .site-title, #site-title a, .site-title a, .entry-title, .entry-title a, h1, h2, h3, h4, h5, h6, .widget-title, .elementor-heading-title { font-family: "Roboto"; } button, .button, input, select, textarea, .wp-block-button, .wp-block-button__link { font-family: "Roboto"; } </style> <!-- Fonts Plugin CSS --> <script> var $container = jQuery('#odsc-tickets'); // on any group click jQuery(document).ready(function(){ jQuery('.grouptickets a').on('click', function (event) { // event && event.preventDefault(); jQuery('#ConferenceTrainingPass').removeClass('groups__group--active'); jQuery('#GroupDiscount').addClass('groups__group--active'); var groupId = "448495d313a47d19c70c96abe7e016b1"; var groupId2 = "70a2bbafefa0ca05dbb0ca678b54582c"; jQuery('.group-content[data-group="'+groupId2+'"]').removeClass("group-content--visible"); jQuery('.group-content[data-group="'+groupId+'"]').addClass("group-content--visible"); // click on subtabs if (jQuery('.group-content__subtab__tab--active:visible').length) { jQuery('.group-content__subtab__tab--active:visible').click(); } else { jQuery('.group-content__subtab__tab:visible:eq(0)').click(); } }); }); </script> <script> var $container = jQuery('#odsc-tickets'); // on any group click jQuery(document).ready(function(){ jQuery('.grouptickets a').on('click', function (event) { // event && event.preventDefault(); jQuery('#BusinessPass').removeClass('groups__group--active'); jQuery('#GroupDiscount').addClass('groups__group--active'); var groupId = "448495d313a47d19c70c96abe7e016b1"; var groupId2 = "d7dca3a8be000a1541cd17e46dc4b9ba"; jQuery('.group-content[data-group="'+groupId2+'"]').removeClass("group-content--visible"); jQuery('.group-content[data-group="'+groupId+'"]').addClass("group-content--visible"); // click on subtabs if (jQuery('.group-content__subtab__tab--active:visible').length) { jQuery('.group-content__subtab__tab--active:visible').click(); } else { jQuery('.group-content__subtab__tab:visible:eq(0)').click(); } }); }); jQuery(document).ready(function(){ jQuery('.trainingtickets a').on('click', function (event) { // event && event.preventDefault(); jQuery('#BusinessPass').removeClass('groups__group--active'); jQuery('#ConferenceTrainingPass').addClass('groups__group--active'); jQuery('#GroupDiscount').removeClass('groups__group--active'); var groupId = "70a2bbafefa0ca05dbb0ca678b54582c"; var groupId2 = "d7dca3a8be000a1541cd17e46dc4b9ba"; var groupId3 = "448495d313a47d19c70c96abe7e016b1"; jQuery('.group-content[data-group="'+groupId2+'"]').removeClass("group-content--visible"); jQuery('.group-content[data-group="'+groupId3+'"]').removeClass("group-content--visible"); jQuery('.group-content[data-group="'+groupId+'"]').addClass("group-content--visible"); // click on subtabs if (jQuery('.group-content__subtab__tab--active:visible').length) { jQuery('.group-content__subtab__tab--active:visible').click(); } else { jQuery('.group-content__subtab__tab:visible:eq(0)').click(); } }); }); </script> <script src="https://d3js.org/d3.v3.min.js"></script></head><body data-rsssl=1 class="page-template-default page page-id-67937 page-child parent-pageid-168 cookies-not-set sticky-header header-solid no-section-scroll top-header none be-themes-layout-layout-wide button-shape-rounded opt-panel-cache-on mprm_chrome" data-be-site-layout='layout-wide' data-be-page-template = 'page' ><div class="search-box-wrapper style2-header-search-widget"> <a href="#" class="header-search-form-close"><i class="icon-icon_close font-icon"></i></a><div class="search-box-inner1"><div class="search-box-inner2"></div></div></div><div id="main-wrapper" ><div id="main" class="ajaxable layout-wide" ><header id="header"><div id="header-top-bar"><div id="header-top-bar-wrap" class=" clearfix"><div id="header-top-bar-left"> <span class="top-bar-widgets">Open Data Science Conference</span></div><div id="header-top-bar-right"><div class="topbar-menu-container"><ul id="topbar-menu" class="clearfix"><li id="menu-item-35199" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-35199"><a href="/boston/bootcamp/" title="AI Bootcamp">AI Bootcamp</a></li><li id="menu-item-40495" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-40495"><a href="/boston" title="EAST">EAST</a></li><li id="menu-item-45816" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-45816"><a href="/europe" title="EUROPE">EUROPE</a></li><li id="menu-item-19686" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-19686"><a href="/apac" title="APAC">APAC</a></li><li id="menu-item-633" class="menu-item menu-item-type-post_type menu-item-object-page current-page-ancestor menu-item-633"><a href="https://odsc.com/california/" title="WEST">WEST</a></li><li id="menu-item-52199" class="nomob menu-item menu-item-type-custom menu-item-object-custom menu-item-52199"><a href="https://aiplus.training/" title="Ai+">Ai+</a></li><li id="menu-item-642" class="nomob menu-item menu-item-type-custom menu-item-object-custom menu-item-642"><a href="https://opendatascience.com" title="Blog">Blog</a></li><li id="menu-item-41261" class="nomob menu-item menu-item-type-custom menu-item-object-custom menu-item-41261"><a href="/boston/partner-odsc-east/" title="Partners">Partners</a></li><li id="menu-item-1102" class="nomob menu-item menu-item-type-custom menu-item-object-custom menu-item-1102"><a href="https://jobs.opendatascience.com/" title="Jobs">Jobs</a></li><li id="menu-item-60037" class="nomob menu-item menu-item-type-custom menu-item-object-custom menu-item-60037"><a href="/podcast" title="Podcast">Podcast</a></li><li id="menu-item-644" class="nomob menu-item menu-item-type-custom menu-item-object-custom menu-item-644"><a href="https://opendatascience.com/newsletter/" title="Newsletter">Newsletter</a></li><li id="menu-item-632" class="nomob menu-item menu-item-type-post_type menu-item-object-page menu-item-632"><a href="https://odsc.com/about-open-data-science-conference/" title="About">About</a></li><li class="menu-item menu-item-gtranslate menu-item-has-children notranslate"><a href="#" data-gt-lang="en" class="gt-current-wrapper notranslate"><img src="https://odsc.com/wp-content/plugins/gtranslate/flags/svg/en.svg" width="16" height="16" alt="en" loading="lazy"></a><ul class="dropdown-menu sub-menu"><li class="menu-item menu-item-gtranslate-child"><a href="#" data-gt-lang="en" class="gt-current-lang notranslate"><img src="https://odsc.com/wp-content/plugins/gtranslate/flags/svg/en.svg" width="16" height="16" alt="en" loading="lazy"></a></li><li class="menu-item menu-item-gtranslate-child"><a href="#" data-gt-lang="fr" class="notranslate"><img src="https://odsc.com/wp-content/plugins/gtranslate/flags/svg/fr.svg" width="16" height="16" alt="fr" loading="lazy"></a></li><li class="menu-item menu-item-gtranslate-child"><a href="#" data-gt-lang="de" class="notranslate"><img src="https://odsc.com/wp-content/plugins/gtranslate/flags/svg/de.svg" width="16" height="16" alt="de" loading="lazy"></a></li><li class="menu-item menu-item-gtranslate-child"><a href="#" data-gt-lang="pt" class="notranslate"><img src="https://odsc.com/wp-content/plugins/gtranslate/flags/svg/pt-br.svg" width="16" height="16" alt="pt" loading="lazy"></a></li><li class="menu-item menu-item-gtranslate-child"><a href="#" data-gt-lang="es" class="notranslate"><img src="https://odsc.com/wp-content/plugins/gtranslate/flags/svg/es.svg" width="16" height="16" alt="es" loading="lazy"></a></li></ul></li></ul></div></div></div></div><div id="header-inner-wrap" class=" style6" ><div id="header-wrap" class=" clearfix" data-default-height="66" data-sticky-height="66"><div class="logo"> <a href="https://odsc.com/california"><img class="transparent-logo dark-scheme-logo" src="https://odsc.com/wp-content/uploads/2019/04/NewLogo3.png" alt="Open Data Science Conference" /><img class="transparent-logo light-scheme-logo" src="https://odsc.com/wp-content/uploads/2019/04/NewLogo3.png" alt="Open Data Science Conference" /><img class="normal-logo" src="https://odsc.com/wp-content/uploads/2019/04/NewLogo3.png" alt="Open Data Science Conference" /><img class="sticky-logo" src="https://odsc.com/wp-content/uploads/2019/04/NewLogo3.png" alt="Open Data Science Conference" /></a></div><div id="header-controls-right"><div class="mobile-nav-controller-wrap"><div class="menu-controls mobile-nav-controller" title="Mobile Menu Controller"> <span class="be-mobile-menu-icon "> <span class="hamburger-line-1"></span> <span class="hamburger-line-2"></span> <span class="hamburger-line-3"></span> </span></div></div></div><nav id="navigation-left-side" class="clearfix"><div class="menu"><ul id="left-menu" class="clearfix be-nav-link-effect-1"><li id="menu-item-775" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-has-children menu-item-775"><a href="/california/conference-tracks/" title="Tracks">Tracks</a> <span class="mobile-sub-menu-controller"><i class="icon-icon_plus"></i></span><ul class="sub-menu clearfix"><li id="menu-item-51828" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-51828"><a href="/california/bootcamp/" title="AI Mini-Bootcamp">AI Mini-Bootcamp</a></li><li id="menu-item-20163" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-20163"><a href="/california/training/" title="Hands-on Training">Hands-on Training</a></li><li id="menu-item-114071" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-114071"><a href="/california/conference-tracks#machine" title="Machine Learning">Machine Learning</a></li><li id="menu-item-95282" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-95282"><a href="/california/conference-tracks#genai" title="Generative AI">Generative AI</a></li><li id="menu-item-114069" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-114069"><a href="/california/conference-tracks#agent" title="AI Agents">AI Agents</a></li><li id="menu-item-51823" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-51823"><a href="/california/conference-tracks#llm" title="LLMs and RAG">LLMs and RAG</a></li><li id="menu-item-36346" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-36346"><a href="/california/conference-tracks#llmops" title="LLMOps & MLOps">LLMOps & MLOps</a></li><li id="menu-item-115267" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-115267"><a href="/california/conference-tracks#nlp" title="AI Engineering">AI Engineering</a></li><li id="menu-item-114070" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-114070"><a href="/california/conference-tracks#deep" title="Deep Learning & Multi-Modal">Deep Learning & Multi-Modal</a></li><li id="menu-item-53267" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-53267"><a href="/california/conference-tracks#responsible" title="Responsible AI">Responsible AI</a></li><li id="menu-item-95407" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-95407"><a href="/california/conference-tracks#dataviz" title="Data Visualization & Analysis">Data Visualization & Analysis</a></li><li id="menu-item-110591" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-110591"><a href="/california/conference-tracks#robot" title="AI for Robotics">AI for Robotics</a></li><li id="menu-item-115266" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-115266"><a href="/california/conference-tracks#agent" title="What's Next in AI?">What’s Next in AI?</a></li></ul></li><li id="menu-item-52423" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-has-children menu-item-52423"><a href="/california/bootcamp/" title="AI Bootcamp">AI Bootcamp</a> <span class="mobile-sub-menu-controller"><i class="icon-icon_plus"></i></span><ul class="sub-menu clearfix"><li id="menu-item-52952" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-52952"><a href="/california/bootcamp/#program" title="Program Overview">Program Overview</a></li><li id="menu-item-52951" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-52951"><a href="/california/bootcamp/#instructors" title="Specialization Tracks">Specialization Tracks</a></li></ul></li><li id="menu-item-17367" class="menu-item menu-item-type-post_type menu-item-object-page menu-item-has-children menu-item-17367"><a href="https://odsc.com/california/why-attend/" title="Attend">Attend</a> <span class="mobile-sub-menu-controller"><i class="icon-icon_plus"></i></span><ul class="sub-menu clearfix"><li id="menu-item-104949" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-104949"><a href="/california/virtual/" title="Virtual Conference">Virtual Conference</a></li><li id="menu-item-22862" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-22862"><a href="/california/why-attend#whoattends" title="See Who Attends">See Who Attends</a></li><li id="menu-item-22861" class="menu-item menu-item-type-post_type menu-item-object-page menu-item-22861"><a href="https://odsc.com/california/why-attend/" title="Why Attend West">Why Attend West</a></li><li id="menu-item-112136" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-112136"><a href="/california/boss/" title="Convince your Manager">Convince your Manager</a></li><li id="menu-item-22863" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-22863"><a href="/california/why-attend#attendeeguide" title="Download Attendee Guide">Download Attendee Guide</a></li><li id="menu-item-118356" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-118356"><a href="https://odsc.com/california/startup-showcase/#Dinner" title="Startup Dinner">Startup Dinner</a></li></ul></li><li id="menu-item-41982" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-has-children menu-item-41982"><a href="/california/schedule/" title="Schedule">Schedule</a> <span class="mobile-sub-menu-controller"><i class="icon-icon_plus"></i></span><ul class="sub-menu clearfix"><li id="menu-item-78786" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-78786"><a href="/california/schedule-overview/" title="Schedule Overview">Schedule Overview</a></li><li id="menu-item-119992" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-119992"><a href="/california/virtual/" title="Virtual Schedule">Virtual Schedule</a></li><li id="menu-item-115556" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-115556"><a href="/california/schedule/" title="Full Schedule">Full Schedule</a></li><li id="menu-item-120719" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-120719"><a href="/california/2024-prereqs/" title="Prerequisites">Prerequisites</a></li><li id="menu-item-38497" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-38497"><a href="/california/bootcamp/#program" title="Mini-Bootcamp Program">Mini-Bootcamp Program</a></li><li id="menu-item-119413" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-119413"><a href="/california/hackathon/" title="NVIDIA Hackathons">NVIDIA Hackathons</a></li><li id="menu-item-105661" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-105661"><a href="/california/events-west/" title="Networking +">Networking +</a></li></ul></li><li id="menu-item-110148" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-has-children menu-item-110148"><a href="/california#register" title="Register">Register</a> <span class="mobile-sub-menu-controller"><i class="icon-icon_plus"></i></span><ul class="sub-menu clearfix"><li id="menu-item-110149" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-110149"><a href="/california/bootcamp" title="AI Mini-Bootcamp">AI Mini-Bootcamp</a></li><li id="menu-item-110150" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-110150"><a href="/california#register" title="Conference Pass">Conference Pass</a></li><li id="menu-item-110151" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-110151"><a href="/media" title="Media Pass">Media Pass</a></li><li id="menu-item-113970" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-113970"><a href="/california/discounts/" title="Discounts">Discounts</a></li><li id="menu-item-119056" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-119056"><a href="/california/hackathon/" title="NVIDIA Hackathons">NVIDIA Hackathons</a></li><li id="menu-item-110152" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-110152"><a href="/code-of-conduct/" title="Code of Conduct">Code of Conduct</a></li></ul></li></ul></div></nav><nav id="navigation-right-side" class="clearfix"><div class="menu"><ul id="right-menu" class="clearfix be-nav-link-effect-1"><li id="menu-item-102800" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-has-children menu-item-102800"><a href="/california/speakers" title="Speakers">Speakers</a> <span class="mobile-sub-menu-controller"><i class="icon-icon_plus"></i></span><ul class="sub-menu clearfix"><li id="menu-item-112836" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-112836"><a href="/california/speakers" title="Confirmed Speakers">Confirmed Speakers</a></li><li id="menu-item-52001" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-52001"><a href="/california/call-for-speakers-west/" title="Call for Speakers">Call for Speakers</a></li></ul></li><li id="menu-item-120689" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-has-children menu-item-120689"><a href="/california/expo" title="EXPO">EXPO</a> <span class="mobile-sub-menu-controller"><i class="icon-icon_plus"></i></span><ul class="sub-menu clearfix"><li id="menu-item-111215" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-111215"><a href="/california/partner-odsc-west/#partners" title="2024 Partners">2024 Partners</a></li><li id="menu-item-102047" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-102047"><a href="/california/partner-odsc-west/#contact" title="Partner Inquiry">Partner Inquiry</a></li><li id="menu-item-105620" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-105620"><a href="/california/expo" title="AI Expo">AI Expo</a></li><li id="menu-item-118355" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-118355"><a href="https://odsc.com/california/startup-showcase" title="Startup Showcase - Apply">Startup Showcase – Apply</a></li><li id="menu-item-118354" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-118354"><a href="https://odsc.com/california/startup-showcase/#Dinner" title="Startup Dinner">Startup Dinner</a></li></ul></li><li id="menu-item-112559" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-112559"><a href="/california/training/" title="Training">Training</a></li><li id="menu-item-109931" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-109931"><a href="/california/aix-west" title="Gen Ai X">Gen Ai X</a></li><li id="menu-item-840" class="menu-item menu-item-type-post_type menu-item-object-page menu-item-has-children menu-item-840"><a href="https://odsc.com/california/conference-guide/" title="Info">Info</a> <span class="mobile-sub-menu-controller"><i class="icon-icon_plus"></i></span><ul class="sub-menu clearfix"><li id="menu-item-70833" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-70833"><a href="/california/faq" title="FAQs">FAQs</a></li><li id="menu-item-120720" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-120720"><a href="/california/2024-prereqs/" title="Prerequisites">Prerequisites</a></li><li id="menu-item-113969" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-113969"><a href="/california/discounts/" title="Discounts">Discounts</a></li><li id="menu-item-113891" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-113891"><a href="/california/west-venue-2021/" title="Venue">Venue</a></li><li id="menu-item-66064" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-66064"><a href="/california/west-venue-2021/" title="Hotel Deal">Hotel Deal</a></li><li id="menu-item-57571" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-57571"><a href="/policy/" title="Event Policies">Event Policies</a></li><li id="menu-item-52160" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-52160"><a href="/code-of-conduct/" title="Code of Conduct">Code of Conduct</a></li></ul></li></ul></div></nav></div><div class='header-mobile-navigation clearfix'><div class="mobile-menu left-mobile-menu"><ul id="mobile-menu" class="clearfix"><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-has-children menu-item-775"><a href="/california/conference-tracks/" title="Tracks">Tracks</a> <span class="mobile-sub-menu-controller"><i class="icon-icon_plus"></i></span><ul class="sub-menu clearfix"><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-51828"><a href="/california/bootcamp/" title="AI Mini-Bootcamp">AI Mini-Bootcamp</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-20163"><a href="/california/training/" title="Hands-on Training">Hands-on Training</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-114071"><a href="/california/conference-tracks#machine" title="Machine Learning">Machine Learning</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-95282"><a href="/california/conference-tracks#genai" title="Generative AI">Generative AI</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-114069"><a href="/california/conference-tracks#agent" title="AI Agents">AI Agents</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-51823"><a href="/california/conference-tracks#llm" title="LLMs and RAG">LLMs and RAG</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-36346"><a href="/california/conference-tracks#llmops" title="LLMOps & MLOps">LLMOps & MLOps</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-115267"><a href="/california/conference-tracks#nlp" title="AI Engineering">AI Engineering</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-114070"><a href="/california/conference-tracks#deep" title="Deep Learning & Multi-Modal">Deep Learning & Multi-Modal</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-53267"><a href="/california/conference-tracks#responsible" title="Responsible AI">Responsible AI</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-95407"><a href="/california/conference-tracks#dataviz" title="Data Visualization & Analysis">Data Visualization & Analysis</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-110591"><a href="/california/conference-tracks#robot" title="AI for Robotics">AI for Robotics</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-115266"><a href="/california/conference-tracks#agent" title="What's Next in AI?">What’s Next in AI?</a></li></ul></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-has-children menu-item-52423"><a href="/california/bootcamp/" title="AI Bootcamp">AI Bootcamp</a> <span class="mobile-sub-menu-controller"><i class="icon-icon_plus"></i></span><ul class="sub-menu clearfix"><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-52952"><a href="/california/bootcamp/#program" title="Program Overview">Program Overview</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-52951"><a href="/california/bootcamp/#instructors" title="Specialization Tracks">Specialization Tracks</a></li></ul></li><li class="menu-item menu-item-type-post_type menu-item-object-page menu-item-has-children menu-item-17367"><a href="https://odsc.com/california/why-attend/" title="Attend">Attend</a> <span class="mobile-sub-menu-controller"><i class="icon-icon_plus"></i></span><ul class="sub-menu clearfix"><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-104949"><a href="/california/virtual/" title="Virtual Conference">Virtual Conference</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-22862"><a href="/california/why-attend#whoattends" title="See Who Attends">See Who Attends</a></li><li class="menu-item menu-item-type-post_type menu-item-object-page menu-item-22861"><a href="https://odsc.com/california/why-attend/" title="Why Attend West">Why Attend West</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-112136"><a href="/california/boss/" title="Convince your Manager">Convince your Manager</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-22863"><a href="/california/why-attend#attendeeguide" title="Download Attendee Guide">Download Attendee Guide</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-118356"><a href="https://odsc.com/california/startup-showcase/#Dinner" title="Startup Dinner">Startup Dinner</a></li></ul></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-has-children menu-item-41982"><a href="/california/schedule/" title="Schedule">Schedule</a> <span class="mobile-sub-menu-controller"><i class="icon-icon_plus"></i></span><ul class="sub-menu clearfix"><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-78786"><a href="/california/schedule-overview/" title="Schedule Overview">Schedule Overview</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-119992"><a href="/california/virtual/" title="Virtual Schedule">Virtual Schedule</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-115556"><a href="/california/schedule/" title="Full Schedule">Full Schedule</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-120719"><a href="/california/2024-prereqs/" title="Prerequisites">Prerequisites</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-38497"><a href="/california/bootcamp/#program" title="Mini-Bootcamp Program">Mini-Bootcamp Program</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-119413"><a href="/california/hackathon/" title="NVIDIA Hackathons">NVIDIA Hackathons</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-105661"><a href="/california/events-west/" title="Networking +">Networking +</a></li></ul></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-has-children menu-item-110148"><a href="/california#register" title="Register">Register</a> <span class="mobile-sub-menu-controller"><i class="icon-icon_plus"></i></span><ul class="sub-menu clearfix"><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-110149"><a href="/california/bootcamp" title="AI Mini-Bootcamp">AI Mini-Bootcamp</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-110150"><a href="/california#register" title="Conference Pass">Conference Pass</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-110151"><a href="/media" title="Media Pass">Media Pass</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-113970"><a href="/california/discounts/" title="Discounts">Discounts</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-119056"><a href="/california/hackathon/" title="NVIDIA Hackathons">NVIDIA Hackathons</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-110152"><a href="/code-of-conduct/" title="Code of Conduct">Code of Conduct</a></li></ul></li></ul></div><div class="mobile-menu right-mobile-menu"><ul id="mobile-menu" class="clearfix"><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-has-children menu-item-102800"><a href="/california/speakers" title="Speakers">Speakers</a> <span class="mobile-sub-menu-controller"><i class="icon-icon_plus"></i></span><ul class="sub-menu clearfix"><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-112836"><a href="/california/speakers" title="Confirmed Speakers">Confirmed Speakers</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-52001"><a href="/california/call-for-speakers-west/" title="Call for Speakers">Call for Speakers</a></li></ul></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-has-children menu-item-120689"><a href="/california/expo" title="EXPO">EXPO</a> <span class="mobile-sub-menu-controller"><i class="icon-icon_plus"></i></span><ul class="sub-menu clearfix"><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-111215"><a href="/california/partner-odsc-west/#partners" title="2024 Partners">2024 Partners</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-102047"><a href="/california/partner-odsc-west/#contact" title="Partner Inquiry">Partner Inquiry</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-105620"><a href="/california/expo" title="AI Expo">AI Expo</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-118355"><a href="https://odsc.com/california/startup-showcase" title="Startup Showcase - Apply">Startup Showcase – Apply</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-118354"><a href="https://odsc.com/california/startup-showcase/#Dinner" title="Startup Dinner">Startup Dinner</a></li></ul></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-112559"><a href="/california/training/" title="Training">Training</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-109931"><a href="/california/aix-west" title="Gen Ai X">Gen Ai X</a></li><li class="menu-item menu-item-type-post_type menu-item-object-page menu-item-has-children menu-item-840"><a href="https://odsc.com/california/conference-guide/" title="Info">Info</a> <span class="mobile-sub-menu-controller"><i class="icon-icon_plus"></i></span><ul class="sub-menu clearfix"><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-70833"><a href="/california/faq" title="FAQs">FAQs</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-120720"><a href="/california/2024-prereqs/" title="Prerequisites">Prerequisites</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-113969"><a href="/california/discounts/" title="Discounts">Discounts</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-113891"><a href="/california/west-venue-2021/" title="Venue">Venue</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-66064"><a href="/california/west-venue-2021/" title="Hotel Deal">Hotel Deal</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-57571"><a href="/policy/" title="Event Policies">Event Policies</a></li><li class="menu-item menu-item-type-custom menu-item-object-custom menu-item-52160"><a href="/code-of-conduct/" title="Code of Conduct">Code of Conduct</a></li></ul></li></ul></div></div></div></header> <!-- END HEADER --><div id="content" class="no-sidebar-page"><div id="content-wrap" class="page-builder"><section id="page-content"><div class="clearfix" ><div class="tatsu-gso398apx953rnx1 tatsu-section tatsu-hide-0 tatsu-hide-mobile tatsu-hide-tablet tatsu-hide-laptop tatsu-hide-desktop tatsu-clearfix" data-title="" data-headerscheme="background--dark"><div class='tatsu-section-pad clearfix' data-padding='{"d":"90px 0px 15px 0px"}' data-padding-top='90px'><div class="tatsu-row-wrap tatsu-wrap tatsu-row-one-col tatsu-row-has-one-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-gso398aq218ockc2" ><div class="tatsu-row " ><div class="tatsu-column tatsu-column-no-bg tatsu-one-col tatsu-column-image-none tatsu-column-effect-none tatsu-gso398aq6jbk7yev" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class="tatsu-module tatsu-text-block-wrap tatsu-gso398aqa64tlefl "><div class="tatsu-text-inner tatsu-align-center clearfix" ><style>.tatsu-gso398aqa64tlefl.tatsu-text-block-wrap .tatsu-text-inner{width: 100%;text-align: left;}</style><p>[vc_row][vc_column][vc_column_text]</p></div></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div></div><style>.tatsu-row > .tatsu-gso398aq6jbk7yev.tatsu-column{width: 100%;}.tatsu-gso398aq6jbk7yev.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: none;}.tatsu-gso398aq6jbk7yev > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-gso398aq6jbk7yev > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-gso398aq6jbk7yev > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-gso398aq6jbk7yev > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}</style></div></div></div></div><div class="tatsu-section-background-wrap"><div class = "tatsu-section-background" ></div></div><style>.tatsu-gso398apx953rnx1 .tatsu-section-pad{padding: 90px 0px 15px 0px;}.tatsu-gso398apx953rnx1 > .tatsu-bottom-divider{z-index: 9999;}.tatsu-gso398apx953rnx1 > .tatsu-top-divider{z-index: 9999;}</style></div><div class="tatsu-r1-4Yz6To0 tatsu-section tatsu-bg-overlay tatsu-hide-tablet tatsu-hide-mobile tatsu-hide-laptop tatsu-hide-desktop tatsu-clearfix" data-title="" data-headerscheme="background--dark"><div class='tatsu-section-pad clearfix' data-padding='{"d":"15px 0px 8px 0px"}' data-padding-top='15px'><div class="tatsu-row-wrap tatsu-wrap tatsu-row-one-col tatsu-row-has-one-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-S1eVtzTpjR" ><div class="tatsu-row " ><div class="tatsu-column tatsu-bg-overlay tatsu-one-col tatsu-column-image-none tatsu-column-effect-none tatsu-ByNYMapoR" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class="tatsu-single-image tatsu-module tatsu-image-lazyload tatsu-external-image tatsu-SJOKGTajC " ><div class="tatsu-single-image-inner " style="" ><div class = "tatsu-single-image-padding-wrap" style = "" ></div><img class = "tatsu-gradient-border" data-src = "https://odsc.com/wp-content/uploads/2024/10/WEB-BANER_WEST_W1-1.png" alt =" " src ="data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAQAAAC1HAwCAAAAC0lEQVR42mNkYAAAAAYAAjCB0C8AAAAASUVORK5CYII=" /></div><style>.tatsu-SJOKGTajC .tatsu-single-image-inner{border-style: solid;max-width: 100%;}.tatsu-SJOKGTajC.tatsu-single-image{transform: translate3d(0px,0px, 0);}</style></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div><div class="tatsu-overlay tatsu-column-overlay tatsu-animate-none" ></div></div><style>.tatsu-row > .tatsu-ByNYMapoR.tatsu-column{width: 100%;}.tatsu-ByNYMapoR.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: normal;}.tatsu-ByNYMapoR > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-ByNYMapoR > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-ByNYMapoR > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-ByNYMapoR > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width:1377px) {.tatsu-row > .tatsu-ByNYMapoR.tatsu-column{width: 100%;}}@media only screen and (min-width:768px) and (max-width: 1024px) {.tatsu-row > .tatsu-ByNYMapoR.tatsu-column{width: 100%;}}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-ByNYMapoR.tatsu-column{width: 100%;}}</style></div></div></div></div><div class="tatsu-section-background-wrap"><div class = "tatsu-section-background" ></div></div><div class="tatsu-overlay tatsu-section-overlay"></div><style>.tatsu-r1-4Yz6To0.tatsu-section{background-color: rgba(0,0,0,1);}.tatsu-r1-4Yz6To0 .tatsu-section-pad{padding: 15px 0px 8px 0px;}.tatsu-r1-4Yz6To0 .tatsu-section-offset-wrap{transform: translateY(-0px);}.tatsu-r1-4Yz6To0 > .tatsu-bottom-divider{z-index: 9999;}.tatsu-r1-4Yz6To0 > .tatsu-top-divider{z-index: 9999;}.tatsu-r1-4Yz6To0 .tatsu-section-overlay{mix-blend-mode: normal;}</style></div><div id="aipitch" class="tatsu-hrybf1obqgfnwe4g tatsu-section tatsu-hide-0 tatsu-hide-tablet tatsu-hide-mobile tatsu-hide-laptop tatsu-hide-desktop tatsu-prevent-overflow tatsu-clearfix" data-title="D - Dates" data-headerscheme="background--dark"><div class='tatsu-section-pad clearfix' data-padding='{"d":"10px 0px 10px 7px","m":"0px 0px 0px 7px"}' data-padding-top='0px'><div class="tatsu-row-wrap tatsu-wrap tatsu-row-has-one-half tatsu-row-has-two-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-hrybf1obt11a66py" ><div class="tatsu-row " ><div class="tatsu-column tatsu-bg-overlay tatsu-one-half tatsu-column-align-middle tatsu-column-image-none tatsu-column-effect-none tatsu-hrybf1obvifvqrgb" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class="tatsu-module tatsu-normal-button tatsu-button-wrap align-block block-center tatsu-hrybf1obxd7yb3ad "><a class="tatsu-shortcode x-largebtn tatsu-button left-icon circular tatsu-animate bg-animation-none " href="/california#register" style= "" data-animation="pulse" aria-label="Register now | last chance" data-gdpr-atts={} target="_blank">Register now | last chance</a><style>.tatsu-hrybf1obxd7yb3ad .tatsu-button{background-color: rgba(245,166,35,1);color: rgba(0,0,0,1) ;border-width: 2px;border-color: rgba(255,27,0,1); }.tatsu-hrybf1obxd7yb3ad .tatsu-button:hover{background-color: rgba(126,211,33,1);color: rgba(0,0,0,1) ;border-color: #00bcdd; }.tatsu-hrybf1obxd7yb3ad.tatsu-normal-button{margin: 0px 0px 0px 0px;}.tatsu-hrybf1obxd7yb3ad{border-color: rgba(0,0,0,1); padding: 0px 0px 0px 60px;border-radius: 5px;}@media only screen and (max-width: 767px) {.tatsu-hrybf1obxd7yb3ad.tatsu-normal-button{margin: 0px 0px 0px 32px;}}</style></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div><div class="tatsu-overlay tatsu-column-overlay tatsu-animate-none" ></div></div><style>.tatsu-row > .tatsu-hrybf1obvifvqrgb.tatsu-column{width: 50%;}.tatsu-hrybf1obvifvqrgb.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: normal;}.tatsu-hrybf1obvifvqrgb > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-hrybf1obvifvqrgb > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-hrybf1obvifvqrgb > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-hrybf1obvifvqrgb > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width:1377px) {.tatsu-row > .tatsu-hrybf1obvifvqrgb.tatsu-column{width: 50%;}}@media only screen and (min-width:768px) and (max-width: 1024px) {.tatsu-row > .tatsu-hrybf1obvifvqrgb.tatsu-column{width: 50%;}}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-hrybf1obvifvqrgb.tatsu-column{width: 100%;}}</style></div><div class="tatsu-column tatsu-bg-overlay tatsu-one-half tatsu-column-image-none tatsu-column-effect-none tatsu-hrybf1obyi2ikqeq" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class="be-countdown-wrap tatsu-hrybf1oc00d4aqjz oshine-module clearfix " ><div class="be-countdown clearfix" data-time="2024-10-31 23:59:59"></div><style>.tatsu-hrybf1oc00d4aqjz .countdown-section{color: rgba(245,166,35,1) ;}.tatsu-hrybf1oc00d4aqjz{border-style: solid;border-width: 5px 5px 5px 5px;border-color: rgba(0,0,0,1); border-radius: 0px;box-shadow: 0px 0px 0px 0px rgba(0,0,0,1);}</style></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div><div class="tatsu-overlay tatsu-column-overlay tatsu-animate-none" ></div></div><style>.tatsu-row > .tatsu-hrybf1obyi2ikqeq.tatsu-column{width: 50%;}.tatsu-hrybf1obyi2ikqeq.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: normal;}.tatsu-hrybf1obyi2ikqeq > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-hrybf1obyi2ikqeq > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-hrybf1obyi2ikqeq > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-hrybf1obyi2ikqeq > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width:1377px) {.tatsu-row > .tatsu-hrybf1obyi2ikqeq.tatsu-column{width: 50%;}}@media only screen and (min-width:768px) and (max-width: 1024px) {.tatsu-row > .tatsu-hrybf1obyi2ikqeq.tatsu-column{width: 50%;}}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-hrybf1obyi2ikqeq.tatsu-column{width: 100%;}}</style></div></div></div></div><div class="tatsu-section-background-wrap"><div class = "tatsu-section-background" ></div></div><style>.tatsu-hrybf1obqgfnwe4g.tatsu-section{background-color: rgba(0,0,0,1);}.tatsu-hrybf1obqgfnwe4g .tatsu-section-pad{padding: 10px 0px 10px 7px;}.tatsu-hrybf1obqgfnwe4g > .tatsu-bottom-divider{z-index: 9999;}.tatsu-hrybf1obqgfnwe4g > .tatsu-top-divider{z-index: 9999;}@media only screen and (max-width: 767px) {.tatsu-hrybf1obqgfnwe4g .tatsu-section-pad{padding: 0px 0px 0px 7px;}}</style></div><div class="tatsu-B1dF74Mhn tatsu-section tatsu-bg-overlay tatsu-hide-0 tatsu-hide-desktop tatsu-hide-laptop tatsu-clearfix" data-title="" data-headerscheme="background--dark"><div class='tatsu-section-pad clearfix' data-padding='{"d":"150px 0px 150px 0px","m":"93px 0px 52px 0px"}' data-padding-top='93px'><div class="tatsu-row-wrap tatsu-wrap tatsu-row-one-col tatsu-row-has-one-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-ByxOF7Nz3h" ><div class="tatsu-row " ><div class="tatsu-column tatsu-column-no-bg tatsu-one-col tatsu-column-image-none tatsu-column-effect-none tatsu-rJW_F7EM22" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class="special-heading-wrap style1 oshine-module tatsu-r1GOKmEMh3 tatsu-hide-desktop tatsu-hide-laptop tatsu-hide-tablet tatsu-hide-mobile " ><div class="special-heading align-center"><h1 class="special-h-tag" >ODSC West 2022</h1><div class="sub-title margin-bottom special-subtitle"><h1 style="text-align: center;"><span style="color: #ffffff;"> Schedule</span></h1><p style="text-align: center;"><span style="color: #ffffff; font-size: 14pt;">more sessions added weekly</span></p></div><div class="sep-with-icon-wrap margin-bottom"><span class="sep-with-icon" ></span><i class="sep-icon font-icon oshine_diamond"></i><span class="sep-with-icon" ></span></div></div><style>.tatsu-r1GOKmEMh3 .special-h-tag{color: rgba(255,255,255,1) ;}.tatsu-r1GOKmEMh3 .sep-icon.oshine_diamond{background: #00bcdd;}.tatsu-r1GOKmEMh3 .sep-with-icon{height: 1px;width: 20px;background: #efefef;}.tatsu-r1GOKmEMh3{padding: 27px 0px 0px 0px;}</style></div><div class="tatsu-single-image tatsu-module align-center tatsu-external-image tatsu-BJQuYQ4M32 " ><div class="tatsu-single-image-inner " style="" ><div class = "tatsu-single-image-padding-wrap" style = "" ></div><img class = "tatsu-gradient-border" src = "https://odsc.com/wp-content/uploads/2024/10/HERO-Schedule-WEST2024_HERO_MOB-overlay-1.png" alt =" " /></div><style>.tatsu-BJQuYQ4M32 .tatsu-single-image-inner{border-style: solid;max-width: 94%;}.tatsu-BJQuYQ4M32.tatsu-single-image{transform: translate3d(0px,0px, 0);}.tatsu-BJQuYQ4M32{padding: 100px 0px 0px 90px;}@media only screen and (max-width: 767px) {.tatsu-BJQuYQ4M32{margin: 0px 0px 0px 0px;padding: 0px 0px 0px 0px;}.tatsu-BJQuYQ4M32 .tatsu-single-image-inner{max-width: 100%;}}</style></div><div class="tatsu-single-image tatsu-module align-center tatsu-external-image tatsu-SyIYS4f2h " ><div class="tatsu-single-image-inner " style="" ><div class = "tatsu-single-image-padding-wrap" style = "" ></div><img class = "tatsu-gradient-border" src = "" alt =" " /></div><style>.tatsu-SyIYS4f2h .tatsu-single-image-inner{border-style: solid;max-width: 94%;}.tatsu-SyIYS4f2h.tatsu-single-image{transform: translate3d(0px,0px, 0);}.tatsu-SyIYS4f2h{padding: 100px 0px 0px 90px;}@media only screen and (max-width: 767px) {.tatsu-SyIYS4f2h{margin: 30px 0px 0px 0px;padding: 0px 0px 0px 0px;}.tatsu-SyIYS4f2h .tatsu-single-image-inner{max-width: 63%;}}</style></div><div class="tatsu-module tatsu-normal-button tatsu-button-wrap align-block block-center tatsu-SyNdYmVf2n tatsu-hide-laptop tatsu-hide-desktop "><a class="tatsu-shortcode largebtn tatsu-button left-icon circular bg-animation-none " href="/california/#register" style= "" aria-label="Register now | last chance" data-gdpr-atts={} target="_blank">Register now | last chance</a><style>.tatsu-SyNdYmVf2n .tatsu-button{background-color: rgba(255,169,0,1);color: rgba(0,0,0,1) ;border-width: 2px;}.tatsu-SyNdYmVf2n .tatsu-button:hover{background-color: rgba(0,0,0,1);color: rgba(255,169,0,1) ;border-color: rgba(255,196,0,1); }.tatsu-SyNdYmVf2n{padding: 0px 0px 55px 90px;}@media only screen and (max-width: 767px) {.tatsu-SyNdYmVf2n.tatsu-normal-button{margin: 0px 0px 0px 0px;}.tatsu-SyNdYmVf2n{padding: 0px 0px 0px 0px;}}</style></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div></div><style>.tatsu-row > .tatsu-rJW_F7EM22.tatsu-column{width: 100%;}.tatsu-rJW_F7EM22.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: none;}.tatsu-rJW_F7EM22 > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-rJW_F7EM22 > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-rJW_F7EM22 > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-rJW_F7EM22 > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}</style></div></div></div></div><div class="tatsu-section-background-wrap"><div class = "tatsu-section-background" ></div></div><div class="tatsu-overlay tatsu-section-overlay"></div><style>.tatsu-B1dF74Mhn.tatsu-section{background-color: rgba(0,0,0,1);}.tatsu-B1dF74Mhn .tatsu-section-background{background-image: url(https://odsc.com/wp-content/uploads/2024/07/blue-west-preliminary-schedule-copy.png);background-repeat: no-repeat;background-attachment: scroll;background-position: top center;background-size: cover;}.tatsu-B1dF74Mhn .tatsu-bg-blur{background-repeat: no-repeat;background-attachment: scroll;background-position: top center;background-size: cover;}.tatsu-B1dF74Mhn .tatsu-section-pad{padding: 150px 0px 150px 0px;}.tatsu-B1dF74Mhn > .tatsu-bottom-divider{z-index: 9999;}.tatsu-B1dF74Mhn > .tatsu-top-divider{z-index: 9999;}.tatsu-B1dF74Mhn .tatsu-section-overlay{mix-blend-mode: normal;}@media only screen and (max-width: 767px) {.tatsu-B1dF74Mhn .tatsu-section-background{background-position: top center;}.tatsu-B1dF74Mhn .tatsu-bg-blur{background-position: top center;}.tatsu-B1dF74Mhn .tatsu-section-pad{padding: 93px 0px 52px 0px;}}</style></div><div class="tatsu-ht4vzmk5b6aw19b2 tatsu-section tatsu-bg-overlay tatsu-hide-laptop tatsu-hide-desktop tatsu-clearfix" data-title="M - banner" data-headerscheme="background--dark"><div class='tatsu-section-pad clearfix' data-padding='{"d":"90px 0px 90px 0px","m":"15px 0px 15px 0px"}' data-padding-top='15px'><div class="tatsu-row-wrap tatsu-wrap tatsu-row-one-col tatsu-row-has-one-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-ht4vzmk5ez7pxt2" ><div class="tatsu-row " ><div class="tatsu-column tatsu-bg-overlay tatsu-one-col tatsu-column-image-none tatsu-column-effect-none tatsu-ht4vzmk5ji70ife9" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class="tatsu-single-image tatsu-module align-center tatsu-image-lazyload tatsu-external-image tatsu-ht4vzmk6dfbini96 tatsu-hide-laptop tatsu-hide-desktop" ><div class="tatsu-single-image-inner " style="" ><div class = "tatsu-single-image-padding-wrap" style = "" ></div><a href = "/california#registration" ><img class = "tatsu-gradient-border" data-src = "https://odsc.com/wp-content/uploads/2024/10/WEB-BANER_WEST2024_W1-mob-320x180_-1.png" alt =" " src ="data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAQAAAC1HAwCAAAAC0lEQVR42mNkYAAAAAYAAjCB0C8AAAAASUVORK5CYII=" /></a></div><style>.tatsu-ht4vzmk6dfbini96 .tatsu-single-image-inner{border-style: solid;max-width: 80%;border-radius: 30px;}.tatsu-ht4vzmk6dfbini96.tatsu-single-image{transform: translate3d(0px,0px, 0);}.tatsu-ht4vzmk6dfbini96{padding: 30px 0px 0px 0px;}@media only screen and (max-width: 767px) {.tatsu-ht4vzmk6dfbini96{margin: 0px 0px 30px 0px;padding: 0px 0px 0px 0px;}.tatsu-ht4vzmk6dfbini96 .tatsu-single-image-inner{max-width: 100%;}}</style></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div><div class="tatsu-overlay tatsu-column-overlay tatsu-animate-none" ></div></div><style>.tatsu-row > .tatsu-ht4vzmk5ji70ife9.tatsu-column{width: 100%;}.tatsu-ht4vzmk5ji70ife9.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: normal;}.tatsu-ht4vzmk5ji70ife9 > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-ht4vzmk5ji70ife9 > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-ht4vzmk5ji70ife9 > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-ht4vzmk5ji70ife9 > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width:1377px) {.tatsu-row > .tatsu-ht4vzmk5ji70ife9.tatsu-column{width: 100%;}}@media only screen and (min-width:768px) and (max-width: 1024px) {.tatsu-row > .tatsu-ht4vzmk5ji70ife9.tatsu-column{width: 100%;}}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-ht4vzmk5ji70ife9.tatsu-column{width: 100%;}}</style></div></div></div></div><div class="tatsu-section-background-wrap"><div class = "tatsu-section-background" ></div></div><div class="tatsu-overlay tatsu-section-overlay"></div><style>.tatsu-ht4vzmk5b6aw19b2.tatsu-section{background-color: rgba(0,0,0,1);}.tatsu-ht4vzmk5b6aw19b2 .tatsu-section-pad{padding: 90px 0px 90px 0px;}.tatsu-ht4vzmk5b6aw19b2 .tatsu-section-offset-wrap{transform: translateY(-0px);}.tatsu-ht4vzmk5b6aw19b2 > .tatsu-bottom-divider{z-index: 9999;}.tatsu-ht4vzmk5b6aw19b2 > .tatsu-top-divider{z-index: 9999;}.tatsu-ht4vzmk5b6aw19b2 .tatsu-section-overlay{mix-blend-mode: normal;}@media only screen and (max-width: 767px) {.tatsu-ht4vzmk5b6aw19b2 .tatsu-section-pad{padding: 15px 0px 15px 0px;}}</style></div><div class="tatsu-h5nsgn6eou24fx92 tatsu-section tatsu-hide-0 tatsu-hide-desktop tatsu-hide-laptop tatsu-hide-tablet tatsu-hide-mobile tatsu-clearfix" data-title="" data-headerscheme="background--dark"><div class='tatsu-section-pad clearfix' data-padding='{"d":"20px 0px 0px 0px"}' data-padding-top='20px'><div class="tatsu-row-wrap tatsu-wrap tatsu-row-one-col tatsu-row-has-one-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-h5nsgn6et8ew5wf4" ><div class="tatsu-row " ><div class="tatsu-column tatsu-column-no-bg tatsu-one-col tatsu-column-image-none tatsu-column-effect-none tatsu-h5nsgn6ezgd7dils" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class="tatsu-single-image tatsu-module tatsu-image-lazyload tatsu-external-image tatsu-h5nsgn6g5k5o5jh9 " ><div class="tatsu-single-image-inner " style="" ><div class = "tatsu-single-image-padding-wrap" style = "" ></div><a href = "/california/#register" target = "_blank" ><img class = "tatsu-gradient-border" data-src = "https://odsc.com/wp-content/uploads/2022/10/WEST_1440x220_WEBSITE_AI_SELLINGOUTSOON-1.png" alt =" " src ="data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAQAAAC1HAwCAAAAC0lEQVR42mNkYAAAAAYAAjCB0C8AAAAASUVORK5CYII=" /></a></div><style>.tatsu-h5nsgn6g5k5o5jh9 .tatsu-single-image-inner{border-style: solid;max-width: 100%;}.tatsu-h5nsgn6g5k5o5jh9.tatsu-single-image{transform: translate3d(0px,0px, 0);}</style></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div></div><style>.tatsu-row > .tatsu-h5nsgn6ezgd7dils.tatsu-column{width: 100%;}.tatsu-h5nsgn6ezgd7dils.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: none;}.tatsu-h5nsgn6ezgd7dils > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-h5nsgn6ezgd7dils > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-h5nsgn6ezgd7dils > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-h5nsgn6ezgd7dils > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}</style></div></div></div></div><div class="tatsu-section-background-wrap"><div class = "tatsu-section-background" ></div></div><style>.tatsu-h5nsgn6eou24fx92 .tatsu-section-pad{padding: 20px 0px 0px 0px;}.tatsu-h5nsgn6eou24fx92 > .tatsu-bottom-divider{z-index: 9999;}.tatsu-h5nsgn6eou24fx92 > .tatsu-top-divider{z-index: 9999;}</style></div><div class="tatsu-B1-VbY3VAO tatsu-section tatsu-bg-overlay tatsu-hide-desktop tatsu-hide-laptop tatsu-hide-tablet tatsu-hide-mobile tatsu-clearfix" data-title="" data-headerscheme="background--dark"><div class='tatsu-section-pad clearfix' data-padding='{"d":"59px 0px 30px 0px"}' data-padding-top='59px'><div class="tatsu-row-wrap tatsu-wrap tatsu-row-one-col tatsu-row-has-one-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-HklmCGBFZs" ><div class="tatsu-row " ><div class="tatsu-column tatsu-bg-overlay tatsu-one-col tatsu-column-image-none tatsu-column-effect-none tatsu-rk70MBKbo" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class="tatsu-module tatsu-text-block-wrap tatsu-ByfegLN29 "><div class="tatsu-text-inner tatsu-align-center clearfix" ><style>.tatsu-ByfegLN29.tatsu-text-block-wrap .tatsu-text-inner{width: 80%;text-align: left;}</style><h6 style="text-align: center;"><strong>Please Note: In-Person attendees will have access to virtual sessions. If you have a virtual pass, please note that we will not live-stream any in-person sessions. Only virtual sessions will be recorded. The schedule overview is available <a href="https://odsc.com/california/schedule-overview/">HERE</a>.</strong></h6></div></div><div class="tatsu-module tatsu-text-block-wrap tatsu-SysGwvw-i "><div class="tatsu-text-inner tatsu-align-center clearfix" ><style>.tatsu-SysGwvw-i.tatsu-text-block-wrap .tatsu-text-inner{width: 78%;text-align: left;color: rgba(0,0,0,1) ;}.tatsu-SysGwvw-i .tatsu-text-inner *{color: rgba(0,0,0,1) ;}</style><h3 style="text-align: center;">50+ Sessions Added</h3><h4 style="text-align: center;">100+ more coming soon.</h4></div></div><div class="tatsu-single-image tatsu-module tatsu-image-lazyload tatsu-external-image tatsu-hgrt35pjtirrixb " ><div class="tatsu-single-image-inner " style="" ><div class = "tatsu-single-image-padding-wrap" style = "" ></div><a href = "/california#register" target = "_blank" ><img class = "tatsu-gradient-border" data-src = "https://odsc.com/wp-content/uploads/2023/08/WEST_WEB_BTS-1.png" alt =" " src ="data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAQAAAC1HAwCAAAAC0lEQVR42mNkYAAAAAYAAjCB0C8AAAAASUVORK5CYII=" /></a></div><style>.tatsu-hgrt35pjtirrixb{margin: 0px 0px 0px 0px;}.tatsu-hgrt35pjtirrixb .tatsu-single-image-inner{border-style: solid;max-width: 100%;}.tatsu-hgrt35pjtirrixb.tatsu-single-image{transform: translate3d(0px,0px, 0);}</style></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div><div class="tatsu-overlay tatsu-column-overlay tatsu-animate-none" ></div></div><style>.tatsu-row > .tatsu-rk70MBKbo.tatsu-column{width: 100%;}.tatsu-rk70MBKbo.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: normal;}.tatsu-rk70MBKbo > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-rk70MBKbo > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-rk70MBKbo > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-rk70MBKbo > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width:1377px) {.tatsu-row > .tatsu-rk70MBKbo.tatsu-column{width: 100%;}}@media only screen and (min-width:768px) and (max-width: 1024px) {.tatsu-row > .tatsu-rk70MBKbo.tatsu-column{width: 100%;}}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-rk70MBKbo.tatsu-column{width: 100%;}}</style></div></div></div></div><div class="tatsu-section-background-wrap"><div class = "tatsu-section-background" ></div></div><div class="tatsu-overlay tatsu-section-overlay"></div><style>.tatsu-B1-VbY3VAO .tatsu-section-pad{padding: 59px 0px 30px 0px;}.tatsu-B1-VbY3VAO .tatsu-section-offset-wrap{transform: translateY(-0px);}.tatsu-B1-VbY3VAO > .tatsu-bottom-divider{z-index: 9999;}.tatsu-B1-VbY3VAO > .tatsu-top-divider{z-index: 9999;}.tatsu-B1-VbY3VAO .tatsu-section-overlay{mix-blend-mode: normal;}</style></div><div class="tatsu-BJoM6qD2h tatsu-section tatsu-bg-overlay tatsu-hide-desktop tatsu-hide-laptop tatsu-hide-tablet tatsu-hide-mobile tatsu-clearfix" data-title="" data-headerscheme="background--dark"><div class='tatsu-section-pad clearfix' data-padding='{"d":"90px 0px 0px 0px"}' data-padding-top='90px'><div class="tatsu-row-wrap tatsu-wrap tatsu-row-one-col tatsu-row-has-one-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-rkxjz69wnh" ><div class="tatsu-row " ><div class="tatsu-column tatsu-bg-overlay tatsu-one-col tatsu-column-image-none tatsu-column-effect-none tatsu-ryboG6cw22" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class="tatsu-module tatsu-text-block-wrap tatsu-BJMozaqP23 "><div class="tatsu-text-inner tatsu-align-center clearfix" ><style>.tatsu-BJMozaqP23.tatsu-text-block-wrap .tatsu-text-inner{width: 100%;text-align: left;background-color: rgba(0,0,0,1);}</style><p style="text-align: center;"><span style="color: #ffffff; font-size: 24pt;">Bootcamp/Pre-Bootcamp</span></p></div></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div><div class="tatsu-overlay tatsu-column-overlay tatsu-animate-none" ></div></div><style>.tatsu-row > .tatsu-ryboG6cw22.tatsu-column{width: 100%;}.tatsu-ryboG6cw22.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: normal;}.tatsu-ryboG6cw22 > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-ryboG6cw22 > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-ryboG6cw22 > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-ryboG6cw22 > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width:1377px) {.tatsu-row > .tatsu-ryboG6cw22.tatsu-column{width: 100%;}}@media only screen and (min-width:768px) and (max-width: 1024px) {.tatsu-row > .tatsu-ryboG6cw22.tatsu-column{width: 100%;}}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-ryboG6cw22.tatsu-column{width: 100%;}}</style></div></div></div><div class="tatsu-row-wrap tatsu-wrap tatsu-row-has-one-half tatsu-row-has-two-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-r1XiGTqPhh" ><div class="tatsu-row " ><div class="tatsu-column tatsu-bg-overlay tatsu-one-half tatsu-column-image-none tatsu-column-effect-none tatsu-B1Nifp5D23" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class = "tatsu-module tatsu-icon_card tatsu-SkHoM65wn3 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-SkHoM65wn3 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-SkHoM65wn3 .tatsu-icon_card-title, .tatsu-SkHoM65wn3 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-SkHoM65wn3 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-SkHoM65wn3.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > ODSC Instructor </a></div><div class = "tatsu-icon_card-caption body"><p>Pre-Botocamp: Introduction to Data Course</p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-SyIiGa9vnn tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-SyIiGa9vnn .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-SyIiGa9vnn .tatsu-icon_card-title, .tatsu-SyIiGa9vnn .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-SyIiGa9vnn .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-SyIiGa9vnn.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > ODSC Instructor </a></div><div class = "tatsu-icon_card-caption body"><p>Pre-Bootcamp: Introduction to Programming with Python Course</p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-HJPofTqw32 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-HJPofTqw32 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-HJPofTqw32 .tatsu-icon_card-title, .tatsu-HJPofTqw32 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-HJPofTqw32 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-HJPofTqw32.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > ODSC Instructor </a></div><div class = "tatsu-icon_card-caption body"><p>Pre-Bootcamp: Data Wrangling with Python Course</p></div></div></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div><div class="tatsu-overlay tatsu-column-overlay tatsu-animate-none" ></div></div><style>.tatsu-row > .tatsu-B1Nifp5D23.tatsu-column{width: 50%;}.tatsu-B1Nifp5D23.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: normal;}.tatsu-B1Nifp5D23 > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-B1Nifp5D23 > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-B1Nifp5D23 > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-B1Nifp5D23 > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width:1377px) {.tatsu-row > .tatsu-B1Nifp5D23.tatsu-column{width: 50%;}}@media only screen and (min-width:768px) and (max-width: 1024px) {.tatsu-row > .tatsu-B1Nifp5D23.tatsu-column{width: 50%;}}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-B1Nifp5D23.tatsu-column{width: 100%;}}</style></div><div class="tatsu-column tatsu-bg-overlay tatsu-one-half tatsu-column-image-none tatsu-column-effect-none tatsu-rk9oMp9vhn" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class = "tatsu-module tatsu-icon_card tatsu-H1ijM65Pnh tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-H1ijM65Pnh .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-H1ijM65Pnh .tatsu-icon_card-title, .tatsu-H1ijM65Pnh .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-H1ijM65Pnh .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-H1ijM65Pnh.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > ODSC Instructor </a></div><div class = "tatsu-icon_card-caption body"><p>Pre-Bootcamp: Introduction to SQL Course</p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-BknjG69w23 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-BknjG69w23 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-BknjG69w23 .tatsu-icon_card-title, .tatsu-BknjG69w23 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-BknjG69w23 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-BknjG69w23.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > ODSC Instructor </a></div><div class = "tatsu-icon_card-caption body"><p>Pre-Bootcamp: Introduction to AI Course</p></div></div></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div><div class="tatsu-overlay tatsu-column-overlay tatsu-animate-none" ></div></div><style>.tatsu-row > .tatsu-rk9oMp9vhn.tatsu-column{width: 50%;}.tatsu-rk9oMp9vhn.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: normal;}.tatsu-rk9oMp9vhn > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-rk9oMp9vhn > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-rk9oMp9vhn > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-rk9oMp9vhn > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width:1377px) {.tatsu-row > .tatsu-rk9oMp9vhn.tatsu-column{width: 50%;}}@media only screen and (min-width:768px) and (max-width: 1024px) {.tatsu-row > .tatsu-rk9oMp9vhn.tatsu-column{width: 50%;}}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-rk9oMp9vhn.tatsu-column{width: 100%;}}</style></div></div></div><div class="tatsu-row-wrap tatsu-wrap tatsu-row-one-col tatsu-row-has-one-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-H1ggiMacDhn" ><div class="tatsu-row " ><div class="tatsu-column tatsu-bg-overlay tatsu-column-empty tatsu-one-col tatsu-column-image-none tatsu-column-effect-none tatsu-HyZeoG65P3n" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div><div class="tatsu-overlay tatsu-column-overlay tatsu-animate-none" ></div></div><style>.tatsu-row > .tatsu-HyZeoG65P3n.tatsu-column{width: 100%;}.tatsu-HyZeoG65P3n.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: normal;}.tatsu-HyZeoG65P3n > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-HyZeoG65P3n > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-HyZeoG65P3n > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-HyZeoG65P3n > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width:1377px) {.tatsu-row > .tatsu-HyZeoG65P3n.tatsu-column{width: 100%;}}@media only screen and (min-width:768px) and (max-width: 1024px) {.tatsu-row > .tatsu-HyZeoG65P3n.tatsu-column{width: 100%;}}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-HyZeoG65P3n.tatsu-column{width: 100%;}}</style></div></div></div></div><div class="tatsu-section-background-wrap"><div class = "tatsu-section-background" ></div></div><div class="tatsu-overlay tatsu-section-overlay"></div><style>.tatsu-BJoM6qD2h.tatsu-section{background-color: rgba(0,0,0,1);}.tatsu-BJoM6qD2h .tatsu-section-pad{padding: 90px 0px 0px 0px;}.tatsu-BJoM6qD2h .tatsu-section-offset-wrap{transform: translateY(-0px);}.tatsu-BJoM6qD2h > .tatsu-bottom-divider{z-index: 9999;}.tatsu-BJoM6qD2h > .tatsu-top-divider{z-index: 9999;}.tatsu-BJoM6qD2h .tatsu-section-overlay{mix-blend-mode: normal;}</style></div><div class="tatsu-HJqanqwh3 tatsu-section tatsu-bg-overlay tatsu-clearfix" data-title="Preliminary Schedule" data-headerscheme="background--dark"><div class='tatsu-section-pad clearfix' data-padding='{"d":"0px 0px 30px 0px"}' data-padding-top='0px'><div class="tatsu-row-wrap tatsu-wrap tatsu-row-one-col tatsu-row-has-one-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-r1xcahqPhn" ><style>.tatsu-r1xcahqPhn > .tatsu-row{margin-top: 0px;}.tatsu-r1xcahqPhn.tatsu-row-wrap > .tatsu-row{margin-bottom: -35px;}</style><div class="tatsu-row " ><div class="tatsu-column tatsu-bg-overlay tatsu-one-col tatsu-column-image-none tatsu-column-effect-none tatsu-ByZ9T2cD22" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class="tatsu-module tatsu-text-block-wrap tatsu-Hk79T2cw23 "><div class="tatsu-text-inner tatsu-align-center clearfix" ><style>.tatsu-Hk79T2cw23.tatsu-text-block-wrap .tatsu-text-inner{width: 78%;text-align: left;color: rgba(255,255,255,1) ;}.tatsu-Hk79T2cw23.tatsu-text-block-wrap{margin: 15px 0px 0px 0px;}.tatsu-Hk79T2cw23 .tatsu-text-inner *{color: rgba(255,255,255,1) ;}</style><h3 style="text-align: center;">150+ Sessions Added</h3></div></div><div class="tatsu-single-image tatsu-module tatsu-image-lazyload tatsu-external-image tatsu-ByN5p3qPnh tatsu-hide-tablet tatsu-hide-mobile tatsu-hide-laptop tatsu-hide-desktop" ><div class="tatsu-single-image-inner " style="" ><div class = "tatsu-single-image-padding-wrap" style = "" ></div><a href = "/california#register" target = "_blank" ><img class = "tatsu-gradient-border" data-src = "https://odsc.com/wp-content/uploads/2023/09/WEST_WEB_40RE-1.png" alt =" " src ="data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAQAAAC1HAwCAAAAC0lEQVR42mNkYAAAAAYAAjCB0C8AAAAASUVORK5CYII=" /></a></div><style>.tatsu-ByN5p3qPnh{margin: 0px 0px 0px 0px;}.tatsu-ByN5p3qPnh .tatsu-single-image-inner{border-style: solid;max-width: 100%;}.tatsu-ByN5p3qPnh.tatsu-single-image{transform: translate3d(0px,0px, 0);}</style></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div><div class="tatsu-overlay tatsu-column-overlay tatsu-animate-none" ></div></div><style>.tatsu-row > .tatsu-ByZ9T2cD22.tatsu-column{width: 100%;}.tatsu-ByZ9T2cD22.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: normal;}.tatsu-ByZ9T2cD22 > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-ByZ9T2cD22 > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-ByZ9T2cD22 > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-ByZ9T2cD22 > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width:1377px) {.tatsu-row > .tatsu-ByZ9T2cD22.tatsu-column{width: 100%;}}@media only screen and (min-width:768px) and (max-width: 1024px) {.tatsu-row > .tatsu-ByZ9T2cD22.tatsu-column{width: 100%;}}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-ByZ9T2cD22.tatsu-column{width: 100%;}}</style></div></div></div><div class="tatsu-row-wrap tatsu-wrap tatsu-row-one-col tatsu-row-has-one-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-HJrqa3cv2h" ><div class="tatsu-row " ><div class="tatsu-column tatsu-bg-overlay tatsu-one-col tatsu-column-image-none tatsu-column-effect-none tatsu-SJ8cp25Phh" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class="tatsu-empty-space tatsu-H1nA1dB_R " ><style>.tatsu-H1nA1dB_R.tatsu-empty-space{height: 30px;}</style></div><div class=" tatsu-rJ-CyedBOC tabs oshine-module" ><ul class="clearfix be-tab-header"><li><a id="tatsu-rkC1l_BdC" class="" href="#fragment-1-499863582">TALKS</a></li><li><a id="tatsu-rygAJeOH_C" class="" href="#fragment-2-499863582">Tutorials | Workshops</a></li><li><a id="tatsu-BkelxOB_R" class="" href="#fragment-3-499863582">Training</a></li><li><a id="tatsu-B1SqGl3OR" class="" href="#fragment-4-499863582">Networking +</a></li></ul><div id="fragment-1-499863582" class="clearfix be-tab-content"> <!-- schedule tab start --><div class="schedule-tab-wrapper etn-tab-wrapper schedule-tab-2"><ul class='etn-nav'><li> <a href='#' class='etn-tab-a etn-active' data-id='tab6740cdee4f8a4-0'> <span class='etn-date'>29 Oct</span> <span class=etn-day>Day 1</span> </a></li><li> <a href='#' class='etn-tab-a ' data-id='tab6740cdee4f8a4-1'> <span class='etn-date'>30 Oct</span> <span class=etn-day>Day 2</span> </a></li><li> <a href='#' class='etn-tab-a ' data-id='tab6740cdee4f8a4-2'> <span class='etn-date'>31 Oct</span> <span class=etn-day>Day 3</span> </a></li></ul><div class='etn-tab-content clearfix etn-schedule-wrap'> <!-- start repeatable item --><div class='etn-tab tab-active' data-id='tab6740cdee4f8a4-0'><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>9:00 am - 9:25 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Generative AI</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading active'><p style="width: 70%;float: left;">Ai X KEYNOTE: Searching for Meaning in the Age of AI</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/bryan-mccann'> <img src='https://odsc.com/wp-content/uploads/2024/10/Bryan-McCann_.png' alt='Bryan McCann'> </a></div></div> <i class="etn-icon etn-minus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Bryan McCann</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Co-founder | CTO at You.com</span></div></div><div class="etn-acccordion-contents active"><p> Bryan McCann, you.com’s co-founder and CTO, shares his unconventional journey from studying philosophy and meaning to finding his way to the computer science department at Stanford and working on groundbreaking AI research alongside Richard Socher. Right now, AI is reshaping everything we hold dear - our jobs, creativity, and identities - it's also our greatest source of inspiration. The Age of AI is simultaneously a Renaissance, Enlightenment, Industrial Revolution, and likely source of humanity’s greatest existential crisis. To surmount this, Bryan will discuss how he uses AI responses as new starting points rather than answers, building teams like neural networks optimized for learning, and how the answer to our meaning crisis may to be more like AI. Exploring AI's impact on politics, economics, healthcare, education, and culture, Bryan asserts that we must all take part in authoring humanity's new story — AI can inspire us to become something new, rather than merely replace what we are now.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>9:00 am - 9:25 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">AI Agents</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">ODSC KEYNOTE: The Current and Future State of Autonomous Agents</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/yohei-nakajima'> <img src='https://odsc.com/wp-content/uploads/2024/10/Yohei-Nakajima.png' alt='Yohei Nakajima'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Yohei Nakajima</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Creator of BabyAGI | GP at Untapped Capital </span></div></div><div class="etn-acccordion-contents "><p> In this session, we'll explore the current landscape of autonomous agents and the potential future they hold. We’ll examine how these agents are being deployed today across industries, from customer service to advanced AI research, and the technical challenges they face. As we look to the future, we'll discuss innovations in agent frameworks, task management, and self-improvement, including how agents might evolve to collaborate autonomously on complex, multi-step projects. The discussion will also touch on ethical considerations, agent autonomy limits, and the emerging possibilities of a world where machines manage entire systems with minimal human input.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>9:30 am - 10:00 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Generative AI</span> <span class="firstfocus">AI Agents</span> <span class="secfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Compound AI Systems and the Future of AI Integration</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/lin-qiao'> <img src='https://odsc.com/wp-content/uploads/2024/08/Lin-Qiao-1.png' alt='Lin Qiao'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Lin Qiao</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">CEO and Cofounder | Fireworks AI</span></div></div><div class="etn-acccordion-contents "><p> The genAI landscape is experiencing a paradigm shift from single models to compound AI systems. These combine models across modalities (i.e., text, audio, video) with external knowledge sources into agentic workflows. By drawing on the unique strengths of different models, and providing access to realtime databases and knowledge sources, we can overcome the finite nature of the single-model approach and the probabilistic limitations of genAI. In this talk, Lin Qiao of Fireworks AI will discuss why compound AI systems are the future, the tools necessary to build them, and distinct design approaches and their tradeoffs. Single models have limited capabilities in problem solving, and app developers need access to compound AI systems to drive innovation with high quality. The presentation will dive into the design of a compound AI system that can automatically decompose a complex business task into multiple steps, accessing multiple models across many modalities (text, audio, image and more), retrievers, and external tools. Attendees will learn about the developer toolkit for fast iteration and deployment of compound AI systems. Key Takeaways: Understanding compound AI systems: Attendees will learn what compound AI systems are, and how they differ from single-model approaches. Designing and optimizing compound AI systems: Attendees will understand the best practices and strategies for designing and optimizing compound AI systems, as well as which platforms and tools to work with. Enterprise examples of compound AI systems: Attendees will see examples of enterprises building and deploying production-scale compound AI systems.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>9:35 am - 10:05 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Machine Learning</span> <span class="firstfocus">Intermediate - Advanced</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Explainability Explained: From Beta coefficients to SHAPly Values</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/giorgio-francesco-clauser'> <img src='https://odsc.com/wp-content/uploads/2024/10/Giorgio-Francesco-Clauser_.png' alt='Giorgio Francesco Clauser'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Giorgio Francesco Clauser</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Head of Data at Moneyfarm</span></div></div><div class="etn-acccordion-contents "><p> Explainability Explained: From Beta Coefficients to SHAPly Values"" is a comprehensive exploration of the concept of explainability in machine learning. Beginning with an overview of the regulatory requirements and the fundamental human need for comprehensible models, the talk demonstrates how the challenges around explainability evolved in the context of machine learning systems. A critical distinction between global and local explainability is elucidated, providing attendees with a nuanced understanding of the different levels at which model interpretability can be assessed. Linear models serve as an illustrative starting point, demonstrating how both global and local explanations can be derived from estimated beta coefficients. The discussion then advances to more complex models, such as XGBOOST, highlighting the unique challenges associated with achieving local explainability in these contexts. Attendees gain insight into the intricacies of understanding model decisions on a case-by-case basis, particularly in scenarios where the underlying processes are inherently opaque. The talk culminates in a deep dive into SHAPly values as a powerful mechanism for facilitating local explanations. By leveraging SHAPly values, attendees learn how to uncover insights into individual predictions, enhancing their ability to interpret and trust machine learning models in practical applications. Throughout the presentation, attendees have access to a Python Jupyter Notebook containing meticulously crafted examples, enabling hands-on exploration and reinforcing key concepts with real-world implementations. With a wealth of knowledge and practical tools at their disposal, attendees leave equipped to navigate the complex landscape of machine learning explainability with confidence.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>10:00 am - 10:30 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Responsible AI</span> <span class="firstfocus">ML</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Tackling Socioeconomic Bias in Machine Learning</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/cody-coleman'> <img src='https://odsc.com/wp-content/uploads/2024/05/Cody-Coleman-Headshot.png' alt='Cody Coleman'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Cody Coleman</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">CEO & Co-Founder at Coactive AI</span></div></div><div class="etn-acccordion-contents "><p> Despite the meteoric rise of ML, most commercially available datasets only represent a small fraction of humanity, with a skew towards high-income populations. To combat algorithmic bias, businesses need to train their ML algorithms on datasets that are representative of all the populations that will be affected by AI deployment. Modern companies and HR teams have learned that inclusive, diverse workforces perform better, and it’s time for the ML community to apply the same wisdom to its training data, especially as their products and services start to impact billions of people across emerging economies and developing countries. This presentation will discuss how socioeconomically diverse datasets can help address algorithmic bias, drawing from Cody’s experience co-creating the open-access Dollar Street Dataset (alongside Gapminder, Harvard University, and MLCommons) and his deep academic knowledge in the space.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>10:15 am - 10:45 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Challenges and Considerations in Language Model Evaluation</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/lintang-sutawika'> <img src='https://odsc.com/wp-content/uploads/2024/09/Lintang-Sutawika.png' alt='Lintang Sutawika'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Lintang Sutawika</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Researcher at EleutherAI </span></div></div><div class="etn-acccordion-contents "><p> NLP and Machine Learning rely on benchmarks and evaluation to accurately track progress in the field and assess the efficacy of new models and methodologies. For this reason, good evaluation practices and accurate reporting are crucial. However, language models (LMs) not only inherit the challenges previously faced in benchmarking, but also introduce a slew of novel considerations which can make proper comparison across models difficult, misleading, or near-impossible. In this talk, we will discuss the state of language model evaluation, and highlight current challenges in evaluating language model performance through discussing the various methods of evaluation, tasks and benchmarks commonly associated with evaluating progress in language model research. We will then discuss how these common pitfalls can be addressed and what considerations should be taken to enhance future work.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:00 am - 11:30 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Generative AI</span> <span class="firstfocus">Responsible AI</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Managing the Volatility of AI Applications</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/shreya-rajpal'> <img src='https://odsc.com/wp-content/uploads/2024/07/Shreya-Rajpal-1.png' alt='Shreya Rajpal'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Shreya Rajpal</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">CEO at Guardrails AI</span></div></div><div class="etn-acccordion-contents "><p> In this talk, Shreya will dive into strategies for enhancing the stability and reliability of AI systems. Drawing from a wealth of real-world development experiences, she distinguishes between the tools and techniques that have yielded success and those that have fallen short. With a focus on practical insights, Shreya also casts a forward-looking eye on the horizon of AI tooling, pinpointing emerging trends and anticipated breakthroughs in the field. This talk is tailored for individuals keen on navigating the dynamic terrain of AI application development, providing a balanced mix of retrospective wisdom and proactive predictions on the evolution of open-source AI technologies.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:00 am - 11:30 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="firstfocus">Generative AI</span> <span class="secfocus">Beginner</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Large Model Quality and Evaluation</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/anoop-sinha'> <img src='https://odsc.com/wp-content/uploads/2024/05/Anoop-Sinha.png' alt='Anoop Sinha'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Anoop Sinha</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Research Director, AI & Future Technologies at Google</span></div></div><div class="etn-acccordion-contents "><p> Large model development has required substantial updates to approaches for ML quality and evaluation, given the challenges of the coverage, scale, and wide use cases for what large models are used for. This presentation discusses the challenges of evaluating large models, with case examples from large language models (LLMs) and large multimodal models in particular. Contrasting LLM development with NLP and Search quality evaluation from just a few years ago is instructive. Earlier human-in-the-loop development, benchmarking, and functional testing, all played a significant role in the success of past quality improvement efforts. These techniques are still relevant but are changing. Metrics for evaluating LLMs are evolving rapidly, and there are open questions about what constitutes "quality" for large models. Some of the challenges include developing and using benchmarks that are robust to data contamination, creating human-based evaluation criteria, and assessing the responsibility of LLMs. The presentation also highlighted some promising approaches to evaluating LLMs, including using benchmarks, human input, considering factors like complexity, multilinguality, and responsibility.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:35 am - 12:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">AI Agents</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Simulating Ourselves and Our Societies With Generative Agents</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/joon-sung-park'> <img src='https://odsc.com/wp-content/uploads/2024/05/Joon-Sung-Park.png' alt='Joon Sung Park'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Joon Sung Park</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">CS PhD Candidate at Stanford University</span></div></div><div class="etn-acccordion-contents "><p> Believable proxies of human behavior can empower interactive applications ranging from immersive environments to rehearsal spaces for interpersonal communication to prototyping tools. In this paper, we introduce generative agents--computational software agents that simulate believable human behavior. Generative agents wake up, cook breakfast, and head to work; artists paint, while authors write; they form opinions, notice each other, and initiate conversations; they remember and reflect on days past as they plan the next day. To enable generative agents, we describe an architecture that extends a large language model to store a complete record of the agent's experiences using natural language, synthesize those memories over time into higher-level reflections, and retrieve them dynamically to plan behavior. We instantiate generative agents to populate an interactive sandbox environment inspired by The Sims, where end users can interact with a small town of twenty five agents using natural language. In an evaluation, these generative agents produce believable individual and emergent social behaviors: for example, starting with only a single user-specified notion that one agent wants to throw a Valentine's Day party, the agents autonomously spread invitations to the party over the next two days, make new acquaintances, ask each other out on dates to the party, and coordinate to show up for the party together at the right time. We demonstrate through ablation that the components of our agent architecture--observation, planning, and reflection--each contribute critically to the believability of agent behavior. By fusing large language models with computational, interactive agents, this work introduces architectural and interaction patterns for enabling believable simulations of human behavior.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:35 am - 12:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Generative AI</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Developing and Deploying State-of-the-Art AI: The Power of Wafer-Scale Engines</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/andy-hock-phd'> <img src='https://odsc.com/wp-content/uploads/2024/07/andy-hock.png' alt='Andy Hock, PhD'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Andy Hock, PhD</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Senior Vice President, Product & Strategy at Cerebras Systems</span></div></div><div class="etn-acccordion-contents "><p> The explosion of GenAI has created an insatiable demand for AI compute that runs large workloads, requiring a massive amount of processing power. However, the computational demands of training and running these models have outpaced traditional computing architectures. Enter Wafer-Scale Engines (WSE)—a groundbreaking new type of computer processor designed specifically for generative AI applications. In this talk, Andy Hock, SVP of Product and Strategy at Cerebras Systems, will explore how Wafer-Scale Engines are enabling the generative AI revolution, empowering researchers and businesses to: Build and customize massive generative AI models specific to their business application quickly and easily with complete data security and full ownership of the model. Accelerate training and inference for unprecedented efficiency. Unlock fundamentally new capabilities in generative AI research and development.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:35 am - 12:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Solution Showcase: Streamlining AI Operations: Leveraging Azure AI Studio for GenAIOps</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/ananya-ghosh-chowdhury'> <img src='https://odsc.com/wp-content/uploads/2024/10/Ananya-Ghosh-Chowdhury.png' alt='Ananya Ghosh Chowdhury'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Ananya Ghosh Chowdhury</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Senior Cloud Solution Architect, Data and AI at Microsoft</span></div></div><div class="etn-acccordion-contents "><p> This session provides a comprehensive overview of the capabilities and features of Azure AI Studio, a platform that empowers organizations to build and deploy AI solutions with ease, leveraging cutting-edge AI tools and model of choice. It also highlights the full development lifecycle, from model catalog and prompt flow to GenAIOps along with safe & responsible AI practices. By the end of the demo, attendees will have a clear understanding of how Azure AI Studio can help developers build generative AI solutions and custom copilots.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>12:10 pm - 12:40 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Generative AI</span> <span class="firstfocus">LLMs</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Demystifying LLM Evaluation</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/jason-lopatecki'> <img src='https://odsc.com/wp-content/uploads/2024/05/Jason-Lopatecki-CEO-Arize-AI.png' alt='Jason Lopatecki'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Jason Lopatecki</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">CEO and Co-Founder at Arize AI</span></div></div><div class="etn-acccordion-contents "><p> According to a recent survey, 61.7% of enterprise engineering teams now have or are planning to have a generative AI application this year – and 14.1% are already in production. As companies race to deploy generative AI into their businesses, the need to ensure that LLMs are deployed reliably and responsibly is paramount. LLM evaluation is a key part of this process. Unfortunately, LLM evaluation and performance analysis of models are an area where confusion reigns and the key distinction between LLM model evaluations and LLM system evaluations often gets lost in practice. Even sophisticated teams stare at a sea of leaderboards and libraries and scratch their heads. There is nothing more important to get right, however, than understanding where you can apply an LLM and how well it is doing at a specific task.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>12:10 pm - 12:40 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Data Engineering</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">How Red Hat Modernizes Data Strategy and Customer Master with Syncari – The Future of Master Data Management</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/scott-edmonds'> <img src='https://odsc.com/wp-content/uploads/2024/10/Scott-Edmonds.png' alt='Scott Edmonds'> </a></div><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/marwood-polasek'> <img src='https://odsc.com/wp-content/uploads/2024/10/Marwood-Polasek.png' alt='Marwood Polasek'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Scott Edmonds</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Founding Team, Chief Revenue Officer at Syncari</span></div><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Marwood Polasek</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Senior Principal Business Architect at Red Hat</span></div></div><div class="etn-acccordion-contents "><p> Join this session to learn how Red Hat is advancing its data strategy and mastering customer data management using Syncari’s cutting-edge Autonomous Data Management (ADM) platform. You’ll gain actionable insights into how Red Hat is leveraging modern Master Data Management (MDM) techniques to ensure data accuracy, scalability, and alignment across its organization. Key Learning Outcomes: Red Hat's Modern MDM Journey: Explore the steps Red Hat has taken to transition to a modern data strategy, emphasizing adaptability and scalability in master data management. Reproducible Data Quality Models: Learn how Red Hat uses Syncari's real-time data model to create flexible, reproducible approaches for maintaining data quality and synchronization across domains like sales and marketing. Automated Data Governance: Discover how Red Hat automates governance and consistency checks across critical systems, ensuring data integrity and compliance. Preparing for Future Data Needs: Understand how Red Hat’s approach equips them to handle future challenges, including the integration of AI-driven insights and scaling for growth. This session will provide practical strategies for modernizing data ecosystems and improving data reproducibility, offering insights that can be applied to your organization’s data management practices.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>12:10 pm - 12:40 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Solution Showcase: Building an Open, Governed Lakehouse with Apache Iceberg and Apache Polaris (Incubating)</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/yufei-gu'> <img src='https://odsc.com/wp-content/uploads/2024/10/Yufei-Gu_.png' alt='Yufei Gu'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Yufei Gu</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Senior Software Engineer at Snowflake</span></div></div><div class="etn-acccordion-contents "><p> With the advent of open source table formats, the open data lakehouse architecture brings the performance and atomic transactions of data warehouses to data lakes. Join this session to see how Apache Iceberg and Apache Polaris (Incubating) can enhance an open data strategy with multi-engine interoperability and governance.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>12:20 pm - 12:50 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="firstfocus">Beginner</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Lessons Learned Applying Large Language Models in Healthcare</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/david-talby-phd'> <img src='https://odsc.com/wp-content/uploads/2024/03/David-Talby.png' alt='David Talby, PhD'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>David Talby, PhD</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Chief Technology Officer at John Snow Labs</span></div></div><div class="etn-acccordion-contents "><p> This talk delves into the practical applications of large language models in the healthcare industry, drawing from his extensive experience as the CTO of John Snow Labs, a renowned player in the field of healthcare data solutions. He provides real-world examples and shares his deep knowledge to showcase how these advanced language models are transforming healthcare. Throughout the talk, you'll gain a comprehensive understanding of the challenges and opportunities in leveraging large language models for healthcare tasks. Talby discusses the nuances of working with healthcare data, compliance, and the unique considerations that come with applying state-of-the-art natural language processing techniques in this critical domain.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>1:00 pm - 1:30 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">AI Engineering</span> <span class="firstfocus">ML</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">“Just Do Something with AI”: Bridging the Business Communication Gap for ML Practitioners</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/stephanie-kirmer'> <img src='https://odsc.com/wp-content/uploads/2024/06/Stephanie-Kirmer.png' alt='Stephanie Kirmer'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Stephanie Kirmer</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Senior Machine Learning Engineer at DataGrail</span></div></div><div class="etn-acccordion-contents "><p> Bringing machine learning functions within a business into alignment with the rest of the organization can be a struggle, especially now when AI has advanced so much technologically, and become such an important part of business innovation and success. In this talk, I'll discuss how ML teams and other business areas fail to communicate effectively, why AI is so misunderstood by laypeople, and why this can lead to the failure of business critical AI initiatives. To solve these problems, ML/AI professionals need to take the initiative to communicate about what they do and what value they can contribute. Educated, smart laypeople are frankly confused and misinformed about what AI is, which leads to unachievable expectations and eventual disappointment, even when machine learning is successfully implemented. At the same time, ML/AI professionals need to deeply understand the goals of the business in order to build AI solutions that will meet those goals. AI represents a significant change to business technology, and massive potential for productivity and problem solving, but as practitioners know only too well, it’s not magic. Attendees to this talk will learn how to bridge the gap between business and AI, and will learn key lessons about how to design and deploy AI initiatives to achieve business success.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:00 pm - 2:30 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Solution Showcase: Cleanlab: The standard for building GenAI and RAG systems that actually work</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/curtis-northcutt'> <img src='https://odsc.com/wp-content/uploads/2024/09/Curtis-Northcutt_.png' alt='Curtis Northcutt'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Curtis Northcutt</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">CEO and Co-Founder at Cleanlab</span></div></div><div class="etn-acccordion-contents "><p> In this talk, Curtis covers the theory, algorithms, and math that enables cleanlab, the most popular data centric AI package used by tens of thousands of data scientists to automatically find and fix errors in any ML dataset, to improve the data quality of thousands of different kinds of datasets. Along the way, Curtis shares lessons learned from ten years working with LLMs, ML, and AI solutions. The talk concludes with industry use cases where Cleanlab was used by companies to save several millions of dollars.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:00 pm - 2:30 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Generative AI</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Synthetic Data for Anonymization, Efficiency and Insights</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/tobias-hann-phd'> <img src='https://odsc.com/wp-content/uploads/2024/07/Tobias-Hann.png' alt='Tobias Hann, PhD'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Tobias Hann, PhD</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">CEO at MOSTLY AI</span></div></div><div class="etn-acccordion-contents "><p> The term """"Synthetic Data"""" has been around for many years. But with the rise of powerful Generative AI models this category of data has become an entirely new meaning. This talk provides an overview of what AI generated synthetic data is, how it is used today, and what its benefits, but also its limitations are. We will start with a generic overview, but the focus will be upon tabular (or structured) synthetic data. One key highlight we will discuss is the privacy preserving benefits of working with synthetic data. In the interactive part of the session we will be using an open source tool (DataLLM) to create tabular synthetic data ourselves. This session is worthwhile to attend for anyone interested in ML generated synthetic data and does not require any previous knowledge. Since the focus is on tabular data, the talk will be most relevant for individuals who work with structured data such as in finance, insurance, telecommunications, healthcare, or research.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:00 pm - 2:30 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">MLOps</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Advanced RAG Architectures: Beyond Basic Document Retrieval</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/bill-deweese'> <img src='https://odsc.com/wp-content/uploads/2024/10/Bill-DeWeese.png' alt='Bill DeWeese'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Bill DeWeese</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Co-Founder and CTO at Airia</span></div></div><div class="etn-acccordion-contents "><p> "As organizations rush to implement Retrieval-Augmented Generation (RAG) systems, many encounter challenges that basic implementations fail to address. This session explores advanced architectural patterns and practical strategies for building more reliable and efficient RAG systems that go beyond simple document retrieval. We'll examine key challenges in real-world RAG implementations, including context window optimization, query decomposition, and hybrid retrieval approaches. The presentation will cover three critical areas: multi-vector retrieval strategies for improved accuracy, dynamic context window management for handling complex queries, and recursive retrieval patterns for enhanced reasoning capabilities. Through practical examples using open-source tools, we'll demonstrate how these advanced patterns can significantly improve the quality and reliability of RAG-based applications. Attendees will learn: How to implement hybrid search approaches combining dense and sparse embeddings. Techniques for effective query rewriting and decomposition Strategies for managing and optimizing context windows Methods for evaluating and benchmarking RAG system performance This technical deep dive will provide attendees with actionable insights and reproducible patterns they can immediately apply to enhance their own RAG implementations. The session will conclude with a discussion of common pitfalls and best practices for scaling RAG systems in production environments. We’ll map out the road to AI maturity, providing a technical roadmap that outlines the steps for fully integrating AI into operations. From creating a robust AI infrastructure to managing models across their lifecycle, you’ll learn how to build a foundation that supports long-term AI adoption. Additionally, we’ll explore advanced strategies for user adoption critical to AI success, helping you drive widespread adoption across teams. This session will cover key methods for monitoring resource utilization, deploying AI across distributed environments, and measuring the financial and operational impact of AI to maximize return on investment. This session will provide you with practical techniques and strategies to tackle AI orchestration and model management, helping your organization unlock AI’s full potential. Key takeaways include: Overcoming deployment challenges, with a focus on implementation, security, and AI governance Building a roadmap to AI maturity, from foundational steps to full-scale operational integration Techniques for promoting user adoption and measuring ROI, ensuring that AI delivers measurable value.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:20 pm - 2:50 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Ai for Robotics</span> <span class="firstfocus">Generative AI</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Preference Learning from Minimal Human Feedback for Interactive Autonomy</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/erdem-b%c4%b1y%c4%b1k-phd'> <img src='https://odsc.com/wp-content/uploads/2024/06/Erdem-Biyik.png' alt='Erdem Bıyık, PhD'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Erdem Bıyık, PhD</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Assistant Professor of Computer Science | lead the Learning and Interactive Robot Autonomy Lab (LiraLab) | University of Southern California </span></div></div><div class="etn-acccordion-contents "><p> The lack of large robotics datasets is arguably the most important obstacle in front of robot learning and interactive autonomy. While large pretrained models and algorithms like reinforcement learning from human feedback (RLHF) led to breakthroughs in other domains like natural language processing and computer vision, robotics has not experienced such a significant breakthrough due to the excessive cost of collecting large datasets. In this talk, I will discuss techniques that enable us to train robots from very little human feedback. I will dive into reinforcement learning from human feedback and describe how active learning methods can enable us to make it more data-efficient. I will finally propose an alternative type of human feedback based on language corrections to further improve both data-efficiency and time-efficiency.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:35 pm - 3:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Generative AI</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Getting Started with On-Device AI: Models and Local Inference SDK</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/alex-chen-phd'> <img src='https://odsc.com/wp-content/uploads/2024/10/Alex-Chen-1.png' alt='Alex Chen, PhD'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Alex Chen, PhD</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Founder | CEO at Nexa AI</span></div></div><div class="etn-acccordion-contents "><p> In this session, attendees will learn cutting-edge developments in On-Device AI: AI models and inference solutions. As industries ranging from automotive to healthcare increasingly adopting AI, On-Device AI provides a scalable and secure alternative by running AI solutions locally to addressing key pain points around latency, privacy, and cost efficiency. This session will introduce the pioneering work in on-device AI models, including the compact and high-performance Octopus v2 and Octopus v3 models. These models bring GPT-4o-level function calling accuracy, with a focus on real-time local inference. Additionally, attendees will learn about local on-device inference toolkit (nexa-sdk) and model hub, which provides an easy-to-use platform for finding and deploying ONNX and GGML models directly on devices, supporting multimodal AI tasks (text, image, and audio). Whether you're an AI practitioner or a decision-maker, this session will provide practical insights into implementing On-Device AI and equipping your solution with On-Device AI models.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:35 pm - 3:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Generative AI</span> <span class="firstfocus">LLMs</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">How to build your own AI with open source and Hugging Face</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/jeff-boudier'> <img src='https://odsc.com/wp-content/uploads/2024/06/Jeff-Boudier.png' alt='Jeff Boudier'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Jeff Boudier</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Head of Product at Hugging Face</span></div></div><div class="etn-acccordion-contents "><p> Jeff, Head of Product at Hugging Face, will walk you through the latest and greatest resources, libraries and tools from Hugging Face to build your own AI features and apps with open models and open source. Learn about how to train and deploy LLMs and Generative AI models in your own infrastructure, build your own chatbots with RAG, and take control of your AI destiny with open source.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:35 pm - 3:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Solution Showcase: Efficient AI Scaling: How VESSL AI Enables 100+ LLM Deployments for $10 and Saves $1M Annually</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/jaeman-an'> <img src='https://odsc.com/wp-content/uploads/2024/09/Jaeman-An.png' alt='Jaeman An'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Jaeman An</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Co-founder & CEO at VESSL AI</span></div></div><div class="etn-acccordion-contents "><p> In this demo talk, we’ll explore how VESSL AI empowers enterprises to efficiently scale the deployment of 100+ Large Language Models (LLMs) starting at just $10, helping businesses save over $100K annually in cloud costs. By leveraging Vessl AI’s robust platform, attendees will learn how to streamline LLM deployment, optimize GPU resource allocation, and implement hybrid-cloud strategies to reduce infrastructure spend while maintaining high-performance AI services. We’ll showcase real-world examples from industries like finance, healthcare, and e-commerce, where Vessl AI’s automation and resource management tools have enabled companies to drastically reduce operational costs and deploy complex AI models at scale. This session will highlight key features such as automated scaling, GPU usage optimization, and cost-efficient hybrid-cloud solutions, allowing organizations to manage AI infrastructure more effectively. By the end of the talk, participants will have a clear roadmap for scaling LLMs efficiently, lowering cloud costs, and increasing AI operational efficiency, all while using Vessl AI’s powerful platform to meet enterprise needs. Learning Outcomes: Understand how to deploy and manage 100+ LLMs in production environments, starting at just $10. Learn practical strategies for reducing cloud costs by over $1M annually using VESSL AI’s hybrid-cloud and GPU resource optimization. Gain insights into automating LLM scaling, monitoring, and resource allocation for cost-effective AI operations. Explore real-world industry use cases that demonstrate the efficiency and cost-saving benefits of Vessl AI’s platform across finance, healthcare, and e-commerce. Apply VESSL AI’s features to maximize performance while minimizing infrastructure overhead and cloud spend. Tools and Learning Goals: Tools: VESSL AI, Kubernetes, Cloud Platforms (AWS, GCP, Azure, Oracle, etc), LangChain, vLLM, LiteLLM, Hugging Face Learning Goals: How to utilize VESSL AI’s platform to deploy LLMs from $10 and scale to 100+ models while managing costs. Efficient use of hybrid-cloud strategies to balance on-premise and cloud resources for optimized AI infrastructure. Best practices for leveraging GPU resource management for LLM training and inference in cost-sensitive environments. How to integrate Vessl AI with open-source tools like Hugging Face and LangChain to streamline AI development and deployment.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>3:30 pm - 4:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Data Visualization</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Data Morph: A Cautionary Tale of Summary Statistics</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/stefanie-molin'> <img src='https://odsc.com/wp-content/uploads/2021/04/Stefanie-Molin.png' alt='Stefanie Molin'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Stefanie Molin</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Data Scientist, Software Engineer, Author of Hands-On Data Analysis with Pandas at Bloomberg</span></div></div><div class="etn-acccordion-contents "><p> Statistics do not come intuitively to humans; they always try to find simple ways to describe complex things. Given a complex dataset, they may feel tempted to use simple summary statistics like the mean, median, or standard deviation to describe it. However, these numbers are not a replacement for visualizing the distribution. To illustrate this fact, researchers have generated many datasets that are very different visually, but share the same summary statistics. In this talk, I will discuss """"""""Data Morph"""""""" (https://github.com/stefmolin/data-morph), an open source package that builds on previous research from Autodesk (the """"""""Datasaurus Dozen"""""""" (https://damassets.autodesk.net/content/dam/autodesk/research/publications-assets/pdf/same-stats-different-graphs.pdf)) using simulated annealing to perturb an arbitrary input dataset into a variety of shapes, while preserving the mean, standard deviation, and correlation to multiple decimal points. I will showcase how it works, discuss the challenges faced during development, and explore the limitations of this approach.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>3:30 pm - 4:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Generative AI</span> <span class="firstfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Retrieval Quality Matters for Gen AI Applications</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/frank-liu'> <img src='https://odsc.com/wp-content/uploads/2024/06/Frank-Liu-2.png' alt='Frank Liu'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Frank Liu</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Head of Applied ML at Voyage AI</span></div></div><div class="etn-acccordion-contents "><p> According to McKinsey’s State of AI Report, 63% of all organizations that are using GenAI consider inaccuracy to be a risk factor - the highest percentage of all categories. Retrieval-Augmented Generation (RAG) is a key tool for significantly reducing inaccuracies and grounding LLMs with private knowledge, but its performance is frequently bottlenecked by the retrieval process itself. In this talk, we’ll discuss retrieval quality - one of the key factors in determining whether or not a RAG-based GenAI application can be deployed successfully. We’ll also dive into some real-world applications along with how results were improved by experimenting with and switching to better retrieval strategies.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>3:30 pm - 4:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Solution Showcase: SEETrials: Innovating Clinical Trials with LLMs to Bridge the Gap Between Health Systems and Life Sciences</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/xiaoyan-wang'> <img src='https://odsc.com/wp-content/uploads/2024/09/Xiaoyan-Wang.png' alt='Xiaoyan Wang'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Xiaoyan Wang</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Chief Scientist and Senior Vice President of Life Sciences Solutions at IMO Health</span></div></div><div class="etn-acccordion-contents "><p> Clinical data, with its inherent complexity and variability, presents a challenge for generative AI models, which require extensive fine-tuning and robust data preparation to ensure precise interpretation and reliable outcomes. This presentation will discuss the real-world application of generative AI to streamline and scale the extraction and analysis of safety and efficacy data from clinical trials. Our study introduces "SEETrials," an innovative pipeline leveraging Generative Pre-trained Transformer (GPT) models for automatic extraction of safety and efficacy data from a wide range of clinical trial abstracts. Focused on multiple myeloma (MM) clinical trials, our approach addresses the critical need for timely and precise clinical data in this challenging context. By comparing outcomes across various interventions for relapsed and refractory MM, we demonstrate the utility of our approach in aiding clinical decision-making. This tool not only benefits MM research but also shows promise for other cancer trial types, offering valuable insights for researchers, clinicians, and industry stakeholders aiming to enhance evidence generation and improve patient care.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>4:05 pm - 4:35 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Solution Showcase: Delphina Demo: AI-powered Data Scientist</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/jeremy-hermann'> <img src='https://odsc.com/wp-content/uploads/2024/09/Jeremy-Hermann.png' alt='Jeremy Hermann'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Jeremy Hermann</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Co-founder at Delphina Ai</span></div></div><div class="etn-acccordion-contents "><p> Data science today — to set prices, optimize supply chains, block fraudsters, or personalize a product — requires a ton of painstaking, manual work. In this demo, we’ll show you how Delphina’s AI-powered (junior) data scientist can massively accelerate your team by building ML models end to end — searching through your warehouse to find relevant data, applying transformations, and then training and tuning models. Delphina can then automatically deploy model and supporting data pipelines to production. Data Scientists are in the driver’s seat throughout, with one-click export of modifiable notebooks that show the work.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>4:05 pm - 4:35 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">AI Agents</span> <span class="firstfocus">Machine Learning</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">How AI Agents and Humans Can Work Together to Transform Our Work</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/kian-katanforoosh'> <img src='https://odsc.com/wp-content/uploads/2024/07/Kian-Katanforoosh.png' alt='Kian Katanforoosh'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Kian Katanforoosh</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Founder at Workera</span></div></div><div class="etn-acccordion-contents "><p> As AI continues to evolve, agents are poised to revolutionize how we approach work across various sectors. This talk will delve into the transformative potential of AI agents, with a particular focus on how humans and agents can work together to create a workforce that thrives in the AI era. Attendees will gain insights into practical implementation strategies, best practices for integrating AI agents into existing systems, and emerging trends that are shaping the future of AI in the workplace. As an expert and award-winning AI practitioner teaching the next generation of workers how to leverage this technology, Kian can discuss how real-world multi-agents fit into our evolving world, whether that is through technical skill development, maintaining an agile and resilient workforce, learning how to manage workers and AI agents, and so much more. Attendees will leave Kian’s session with a better understanding of how to leverage AI agents effectively while simultaneously creating a more optimized human workforce that not only solves problems but also continues to drive innovation and success. Practical Implementation Strategies: We will discuss how to successfully integrate AI agents into existing organizational frameworks. This will encompass everything from selecting the right AI technologies to ensuring seamless interaction between AI agents and human workers. Attendees will learn best practices for deploying AI solutions that enhance productivity and efficiency. Best Practices for Integration: Learn how to manage the integration of AI agents within your systems to maximize their benefits. This will cover the technical and managerial aspects of incorporating AI into workflows, including training employees to work alongside AI and addressing potential resistance to change. Emerging Trends: Stay ahead of the curve by understanding the latest developments in AI agent technology. We will explore how advancements in machine learning and data science are driving the evolution of AI agents and their increasing capabilities in solving complex problems. Participants will gain insights into future directions and how to prepare their organizations for upcoming innovations.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>4:05 pm - 5:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Generative AI</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">AI in Industry</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/jayasree-boyapati'> <img src='https://odsc.com/wp-content/uploads/2024/10/Jayasree-Boyapati-1.png' alt='Jayasree Boyapati'> </a></div><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/kerstin-frailey-phd'> <img src='https://odsc.com/wp-content/uploads/2024/10/Kerstin-Frailey-1.png' alt='Kerstin Frailey, PhD'> </a></div><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/rahul-ponnala'> <img src='https://odsc.com/wp-content/uploads/2024/10/Rahul-Ponnala_.png' alt='Rahul Ponnala'> </a></div><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/cal-al-dhubaib'> <img src='https://odsc.com/wp-content/uploads/2021/02/Cal-Al-Dhubaib.png' alt='Cal Al-Dhubaib'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Jayasree Boyapati</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Director in IT at Visa Inc</span></div><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Kerstin Frailey, PhD</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Data Science Manager at Asana</span></div><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Rahul Ponnala</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">CEO at Granica</span></div><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Cal Al-Dhubaib</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Head of AI and Data Science at Further</span></div></div><div class="etn-acccordion-contents "><p> Explore how AI is reshaping various industries in this dynamic panel at the Gen Ai X Summit. Our panelists, leaders from sectors like finance, e-commerce, and collaboration tools, will share their insights on implementing AI to tackle real-world challenges, drive efficiency, and foster innovation. Delve into the opportunities and obstacles faced by organizations adopting AI at scale, and learn how different industries are navigating the rapidly evolving landscape of AI technologies.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>4:40 pm - 5:10 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs & RAG</span> <span class="firstfocus">Deep Learning</span> <span class="secfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Multimodal Retrieval-Augmented Generation (RAG) with Vector Database</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/stefan-webb-phd'> <img src='https://odsc.com/wp-content/uploads/2024/09/Stefan-Webb.png' alt='Stefan Webb, PhD'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Stefan Webb, PhD</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Developer Advocate at Zilliz</span></div></div><div class="etn-acccordion-contents "><p> Multi-modality elevates the capabilities of neural network models to a whole new level. By leveraging contrastive learning and specialized model architectures, we can create a unified vector space for images and text, enhancing multimodal representations. This talk will share insights into building image-text search and Composite Image Retrieval (CIR) using multimodal embeddings and the Milvus vector database, demonstrating how multi-modality unlocks new use cases in Retrieval-Augmented Generation (RAG).</p></div></div></div></div></div></div> <!-- end repeatable item --> <!-- start repeatable item --><div class='etn-tab ' data-id='tab6740cdee4f8a4-1'><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>9:00 am - 9:25 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Responsible AI</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading active'><p style="width: 70%;float: left;">ODSC KEYNOTE: Beyond Models – Applying AI and Data Science Effectively</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/dr-alfred-spector'> <img src='https://odsc.com/wp-content/uploads/2024/09/Alfred-Spector-.png' alt='Dr. Alfred Spector'> </a></div></div> <i class="etn-icon etn-minus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Dr. Alfred Spector</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Visiting Scholar at MIT | Senior Advisor at Blackstone</span></div></div><div class="etn-acccordion-contents active"><p> Applying artificial intelligence and data science effectively requires a considerably broader focus than just data and machine learning. This presentation distills these additional challenges into a rubric and illustrates its application with a number of examples. Beyond the rubric, the presentation also presents useful frameworks for making the complex trade-offs that are present and growing. While the talk should have practical value to those developing, deploying, and regulating AI and DS, it also illustrates contemporary research challenges.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>9:00 am - 9:25 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Tracks</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">ODSC KEYNOTE: Infusing and Scaling Generative AI into Business Differentiation</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/dr-ali-arsanjani'> <img src='https://odsc.com/wp-content/uploads/2024/09/Dr.-Ali-Arsanjani-1.png' alt='Dr. Ali Arsanjani'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Dr. Ali Arsanjani</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Director of Applied AI Engineering | Head of AI Center of Excellence at Google Cloud</span></div></div><div class="etn-acccordion-contents "><p> This presentation focuses on how businesses can leverage generative AI for differentiation. The three key takeaways are: 1) Emerging Trends: Generative AI is rapidly evolving with trends such as large context windows (e.g., Gemini's 2M token context window), multi-modal automation (integrating text, audio, images, video, and code), solution-complete platforms, and advanced agentic architecture (systems that interact with the real world). 2) Maturing and Scaling AI: This involves a six-level journey starting with data preparation (Level 0) and culminating in the deployment of multi-agent systems and LLMOps (Level 6). Each level incorporates increasingly sophisticated techniques like RAG, fine-tuning, and reinforcement learning from human feedback (RLHF). 3) Competitive Differentiation: Key vectors for differentiation include leveraging high-quality data, crafting effective prompts, utilizing RAG for grounding, fine-tuning models for specific tasks, employing multimodal and multi-agent systems, grounding responses in verifiable sources, and implementing robust evaluation and monitoring practices throughout the AI lifecycle.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>9:30 am - 9:55 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="firstfocus">All Tracks</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">ODSC KEYNOTE: From AI to Data Processing: The Next Phase of Accelerated Computing</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/nick-becker'> <img src='https://odsc.com/wp-content/uploads/2024/09/Nick-Becker.png' alt='Nick Becker'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Nick Becker</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Product Leader in GPU-accelerated Data Science at NVIDIA</span></div></div><div class="etn-acccordion-contents "><p> The accelerated computing revolution that ushered in the AI era is now transforming data processing — and just in time. It’s the clear path forward to sustainably processing the massive amount of data enterprises are creating every year and supporting a world in which generative AI assistants enable asking more questions of our data. In this talk, we’ll trace a line through the history and innovations that have led to this point, explore why data processing needs accelerated computing, assess where things stand today, and consider what new experiences this revolution might unlock.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>9:35 am - 10:05 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="firstfocus">Beginner-Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Operationalizing AI Agents in Data Analytics Workflows</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/ines-chami'> <img src='https://odsc.com/wp-content/uploads/2024/10/Ines-Chami.png' alt='Ines Chami'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Ines Chami</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Co-founder and Chief Scientist at Numbers Station AI</span></div></div><div class="etn-acccordion-contents "><p> Integrating Large Language Models (LLMs) into production-level data workflows presents both significant challenges and opportunities. In this talk, we'll introduce Numbers Station, a platform that automates data analytics workflows using LLMs, Retrieval Augmented Generation (RAG) over a Knowledge Layer, and a customizable multi-agent architecture. We'll start by discussing practical use cases for analytics, such as dashboard search, query generation, or automatically summarizing analyses into slide presentations. We'll then delve into the methodologies for deploying LLMs within data analytics workflows, focusing on a detailed case study to build a SQL agent from the ground up. We will cover the architectural considerations necessary to support agent-based analytics, including the role of dynamic control flows and the importance of incorporating business context through a unified Knowledge Layer. This session aims to provide a deep technical insight into transforming theoretical AI frameworks into practical, scalable solutions that advance organizational data capabilities.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>10:05 am - 10:30 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Tracks</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Virtual Keynote Fireside Chat</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/gary-marcus-phd'> <img src='https://odsc.com/wp-content/uploads/2023/08/Gary-Marcus.png' alt='Gary Marcus, PhD'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Gary Marcus, PhD</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Scientist, Best-selling Author, and Serial Entrepreneur </span></div></div><div class="etn-acccordion-contents "><p> Join us for a virtual fireside chat with Gary Marcus, where we'll explore the fascinating world of artificial intelligence. We'll discuss everything from the latest advancements in machine learning to the ethical implications of AI. Don't miss this opportunity to gain insights from one of the leading minds in the field.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>10:30 am - 11:00 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">AI Engineering</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">TRACK KEYNOTE: From ML Engineering to AI Engineering</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/chip-huyen'> <img src='https://odsc.com/wp-content/uploads/2021/07/chip-huyen.png' alt='Chip Huyen'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Chip Huyen</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">VP of AI & OSS at Voltron Data</span></div></div><div class="etn-acccordion-contents "><p> The availability of foundation models has enabled many new applications and lowered barriers to entry for building AI products. But how does building with foundation models differ from traditional ML? This talk explores key shifts, including the increasing challenges of evaluating open-ended outputs, the transition from structured to unstructured data, and the growing integration of product and engineering.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>10:25 am - 10:55 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Data Engineering</span> <span class="firstfocus">Generative AI</span> <span class="secfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Don't Go Over the Deep End: Building an Effective OSS Management Layer for Your Data Lake</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/dr-einat-orr'> <img src='https://odsc.com/wp-content/uploads/2024/07/Dr.-Einat-Orr.png' alt='Dr. Einat Orr'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Dr. Einat Orr</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Co-Founder & CEO at Treeverse</span></div></div><div class="etn-acccordion-contents "><p> Managing a data lake with both structured and unstructured data sometimes feels like diving int a deep abyss, especially for beginners. But it doesn't have to be that way! This talk offers a high-level overview of tools and strategies to enhance data lake manageability—without going off the deep end. We'll start by exploring fundamental challenges, focusing on the different needs of structured versus unstructured data where each requires its own distinct approach. We'll dispel some of the chaos by covering the key components of a robust data lake management architecture, including open table formats, catalogs, and data version control systems. By understanding these components, you'll see how they contribute to an organized data lake environment, helping you avoid feeling like you're constantly treading water. We'll present real life data lake architectures using Databricks, Apache Iceberg, and AWS technologies to show how these components integrate seamlessly in a modern data engineering stack. If time allows, we will conclude with an open discussion, encouraging attendees to share their experiences (read: rants) and challenges, so you can feel less alone in the murky waters of the multi-structure data lake, and come away with practical methods for data lake manageability.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>10:25 am - 10:55 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Generative AI</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Data Science is Dead</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/jure-leskovec'> <img src='https://odsc.com/wp-content/uploads/2024/09/Jure-Leskovec.png' alt='Jure Leskovec'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Jure Leskovec</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Co-founder and Chief Scientist at Kumo.AI</span></div></div><div class="etn-acccordion-contents "><p> Months of effort for a 1% improvement means that traditional data science methods have reached their limits. Intelligent data science breaks through that barrier by reshaping the machine learning lifecycle using AI and unlocking the next echelon of value from enterprise data. Accelerate model creation and performance Eliminate the need for feature engineering Deliver model improvements of up to 75% Intelligent data science delivers answers in hours and allows data scientists to apply their domain expertise to test hypotheses and further refine AI generated models. Used to drive revenue KPIs like engagement, sales, and conversion, learn how and why intelligent data science is upending an entire industry. In this session you’ll learn: What is intelligent data science Best practices and considerations Redefining the strategic role of data scientists and machine learning engineers The new ML lifecycle Measurable outcomes from real use cases This introductory session unveils intelligent data science best practices gathered from data science and AI executives at Pinterest, AirBnB, and LinkedIn, as well as leading academics from Stanford University.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>10:30 am - 11:00 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="firstfocus">Generative AI</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Reinforcement Learning with Human Feedback</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/luis-serrano-phd'> <img src='https://odsc.com/wp-content/uploads/2020/07/Luis-Serrano.png' alt='Luis Serrano, PhD'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Luis Serrano, PhD</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Author of Grokking Machine Learning and Creator of Serrano Academy</span></div></div><div class="etn-acccordion-contents "><p> LLMs have proven to be tremendously successful at generating text. A very important step in their fine-tuning involves humans evaluating the output. In order to improve the model with human feedback, RLHF is a widely used method. In this talk, we'll explore several aspects, including: A brief review of reinforcement learning. How RLHF is used to fine-tune large language models. Proximal Policy Optimization (PPO), the reinforcement learning training technique used for RLHF. Direct Preference Optimization (DPO), an alternate method to fine-tune an LLM with human feedback, which doesn't use RL, and has performed quite well.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:00 am - 11:30 am</span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Scaling GenAI at Cresta</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/tim-shi'> <img src='https://odsc.com/wp-content/uploads/2024/09/Tim-Shi.png' alt='Tim Shi'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Tim Shi</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Co-Founder at Cresta</span></div></div><div class="etn-acccordion-contents "><p> Abstract Coming Soon!</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:00 am - 11:30 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Data Engineering</span> <span class="firstfocus">Beginner - Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">The AI Advantage: Transforming Software Development for Operational Excellence</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/narendra-lakshamana-gowda'> <img src='https://odsc.com/wp-content/uploads/2024/10/Narendra-Lakshamana-Gowda.png' alt='Narendra Lakshamana Gowda'> </a></div><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/balvinder-banjardar'> <img src='https://odsc.com/wp-content/uploads/2024/10/Balvinder-Banjardar_.png' alt='Balvinder Banjardar'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Narendra Lakshamana Gowda</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Senior Engineering Manager & Architect at Walmart Global Tech</span></div><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Balvinder Banjardar</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Director of Software Engineering at Walmart Global Tech</span></div></div><div class="etn-acccordion-contents "><p> In this talk, we delve deeply into the utilization of Artificial Intelligence (AI) on the software development process, meticulously exploring its optimization capabilities across all stages—from initial development to post-production. Our discussion focuses on the practical application of AI in several key areas, including identifying risks in future production deployments, and detecting anomalies in production metrics for early incident mitigation. We present our comprehensive findings on leveraging AI in the core platform development of store systems at a fortune number 1 company, illustrating how AI can be integrated into critical software development process. By drawing attention to the transformative capabilities of AI, we aim to highlight its pivotal role in improving operational efficiency and productivity. This talk proposes a forward-thinking approach to large-scale software development, showcasing how AI can streamline processes, reduce errors, and foster innovation. Join us to gain a thorough understanding of the profound impact of AI on software development at scale, offering an insightful glimpse into the future of the tech industry. This presentation is designed to provide valuable insights for developers, engineers, and tech enthusiasts interested in the future trajectory of software development and the integral role AI plays in it.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:00 am - 11:30 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Solution Showcase: Visualizing AI-Driven Clinical Trial Planning</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/jeremy-zhang-phd'> <img src='https://odsc.com/wp-content/uploads/2024/09/Jeremy-Zhang.png' alt='Jeremy Zhang, PhD'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Jeremy Zhang, PhD</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Head of Advanced Analytics at Gilead Sciences</span></div></div><div class="etn-acccordion-contents "><p> Clinical trials can be costly with low margin for error when it comes to finding the right patients to recruit into trials to ensure trail success. Globally, up to 80% of clinical trials fail due to recruitment challenges. Data-driven and AI-driven techniques to design clinical trials and improve the execution of trials including screening, enrollment, and site selection have become increasingly prevalent in the pharmaceutical industry. At Gilead – AI models are used to enhance decision-making across various aspects of drug development – including the way we design, plan, and execute clinical trials. In this talk – a connection will be made between the data and AI work performed by data scientists and key decisions made by clinical research teams through visualizations in Plotly and Dash.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:00 am - 11:30 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="firstfocus">Intermediate-Advanced</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">DIY LLMs: Rolling an LLM Inference Service from GPUs to o11y</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/charles-frye-phd'> <img src='https://odsc.com/wp-content/uploads/2024/09/Charles-Frye.png' alt='Charles Frye, PhD'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Charles Frye, PhD</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">AI Engineer at Modal Labs</span></div></div><div class="etn-acccordion-contents "><p> Two years after the release of ChatGPT, open models and open source tooling have now made it possible to host your own LLM inference service. From LLaMA and Liger Kernels to Axolotl and vLLM, there's open software solutions for every part of the stack. But where once there was a dearth, we now have an excess. What are all of these tools and which are good for what purposes? In this talk, Modal Labs AI Engineer Charles Frye will walk through the components of a self-hosted LLM inference service, starting at the hardware and passing through internal engineering tools like evals on the way to the application layer. We'll consider some key differentiators between applications that drive engineering constraints and tooling choices: latency-sensitive vs throughput-sensitive workloads, online vs offline evaluations, and more. Attendees will walk away with a mental map of the landscape of LLM tooling, along with recommendations for where to start.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:00 am - 11:30 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Accelerating Data Agents with cuDF Pandas</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/jiwei-liu'> <img src='https://odsc.com/wp-content/uploads/2024/10/Jiwei-Liu.png' alt='Jiwei Liu'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Jiwei Liu</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Kaggle Grandmaster and Data Scientist at NVIDIA </span></div></div><div class="etn-acccordion-contents "><p> In this talk, we will showcase the benefits of accelerating pandas with a custom pandas agent that makes pandas dataframe manipulation conversational. The data agent connects into local data sources and allows analysis with natural language, but for large datasets, the agent's processing time becomes prohibitively long due to slow CPU based Pandas operations. I’ll demonstrate how to leverage RAPIDS cuDF to perform pandas operations generated by pandasAI on a GPU. The operations will seamlessly fall back to Pandas on the CPU when necessary, without requiring any code modifications beyond a simple decorator or additional import. This demo highlights the advancement of data agents and the role they play in protecting data privacy. I’ll connect a local Large Language Model (LLM) to RAPIDS cuDF in pandas accelerator mode, creating an end-to-end, GPU-accelerated data agent that operates entirely on-premises. This solution yields a substantial performance boost while maintaining data security and integrity.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:10 am - 11:40 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Generative AI</span> <span class="firstfocus">ML</span> <span class="secfocus">Beginner-Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Frontiers of Foundation Models for Time Series</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/yan-liu-phd'> <img src='https://odsc.com/wp-content/uploads/2024/07/Yan-Liu.png' alt='Yan Liu, PhD'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Yan Liu, PhD</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Professor at University of Southern California </span></div></div><div class="etn-acccordion-contents "><p> Recent development in deep learning has spurred research advances in time series modeling and analysis. While achieving state-of-the-art results, the best-performing architectures vary highly across applications and domains. Meanwhile, for natural language processing, the Generative Pre-trained Transformer (GPT) has demonstrated impressive performance via training one general-purpose model across various textual datasets. It is intriguing to explore whether GPT-type architectures can be effective for time series, capturing the intrinsic dynamic attributes and leading to significant accuracy improvements. Furthermore, practical applications of time series raise a series of new challenges, such as multi-resolution, multimodal, missing value, distributeness, and interpretability. In this talk, I will discuss the recent development in the area and possible paths to foundation models for time series data. At the end of the talk, I will share my view of future directions for time series research and foundation models. The talk will feature our recent work on ""Tempo: Prompt-based generative pre-trained transformer for time series forecasting"" in ICLR 2024. Learning Objectives and Tools : Time series modeling</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:35 am - 12:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Data Engineering</span> <span class="firstfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Privacy and Security in the Age of Generative AI</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/benjamin-bengfort'> <img src='https://odsc.com/wp-content/uploads/2024/10/Benjamin-Bengfort_.png' alt='Benjamin Bengfort'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Benjamin Bengfort</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">CEO & Co-Founder atRotational Labs</span></div></div><div class="etn-acccordion-contents "><p> From sensitive data leakage to prompt injection and zero-click worms, LLMs and generative models are the new cyber battleground for hackers. As more AI models are deployed in production, data scientists and ML engineers can't ignore these problems. The good news is that we can influence privacy and security in the machine learning lifecycle using data specific techniques. In this talk, we'll review some of the newest security concerns affecting LLMs and deep learning models and learn how to embed privacy into model training with ACLs and differential privacy, secure text generation and function-calling interfaces, and even leverage models to defend other models.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:35 am - 12:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="firstfocus">Generative AI</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Data Exfiltration Attacks in LLM Products</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/arzav-jain'> <img src='https://odsc.com/wp-content/uploads/2024/09/Arzav-Jain.png' alt='Arzav Jain'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Arzav Jain</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Member of Technical Staff at OpenAI</span></div></div><div class="etn-acccordion-contents "><p> From ChatGPT to Gemini, Bing chat to Slack AI, a number of LLM chatbots have been and some still are vulnerable to data exfiltration attacks. In a world where more and more LLM-based applications are being developed, this new risk surface makes it increasingly possible to steal user information unless it's properly addressed. This talk will describe how such attacks are possible via prompt injection and with a detailed walk-through. We will go over possible user data that can be exfiltrated as well as possible mitigation strategies and why some of them don't work. You will walk away with a clear understanding of how such attacks work and how they might impact products that you're building. It's important that we as developers take the responsibility to protect our users as we bring the benefits of such AI applications to them.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:35 am - 12:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Solution Showcase: Mastering Complexity: Optimize your decision making for 500% ROI</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/juan-guzman'> <img src='https://odsc.com/wp-content/uploads/2024/10/Juan-Guzman.png' alt='Juan Guzman'> </a></div><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/jennifer-locke'> <img src='https://odsc.com/wp-content/uploads/2024/03/jennifer_locke-1.png' alt='Jennifer Locke'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Juan Guzman</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Optimization Engineer at Gurobi</span></div><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Jennifer Locke</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Manager – Technical Account Management at Gurobi</span></div></div><div class="etn-acccordion-contents "><p> Businesses focus a lot on forecasts and predictions – trying to get a clearer picture of the future. But even if you had perfect information, the most sprawling and impactful business decisions are much too complex to guarantee optimal outcomes, with millions, billions, or even trillions of trade-offs to consider. See why leading companies use mathematical optimization to solve their most complex real-world business problems. </p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:35 am - 12:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Generative AI</span> <span class="firstfocus">ML</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Data Science in the Age of Generative AI</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/paige-bailey'> <img src='https://odsc.com/wp-content/uploads/2024/07/Paige-Bailey.png' alt='Paige Bailey'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Paige Bailey</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">DevRel Lead, GenAI at Google</span></div></div><div class="etn-acccordion-contents "><p> Generative AI, fueled by powerful models like Gemini, Claude, and GPT-4, is rapidly transforming the data landscape. This talk explores the profound impact of this shift on the role of data scientists and machine learning engineers. We'll discuss how generative AI empowers data scientists with new tools for: * **Data Augmentation and Synthetic Data Generation:** Overcome data scarcity and generate realistic datasets for diverse tasks. * **Automated Feature Engineering and Data Preprocessing:** Simplify complex data pipelines and streamline model development. * **Data Visualization and Exploration:** Gain deeper insights through AI-driven data exploration and visualization. * **Automated Model Building and Optimization:** Leverage AI to rapidly prototype and optimize models for various tasks. * **Explainability and Interpretability:** Decipher the """"black box"""" of AI models and build trust in their predictions. By the end of this session, attendees will understand the transformative potential of Generative AI in data science and gain practical insights into leveraging these new tools for enhanced efficiency, creativity, and impact.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:35 am - 12:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Generative AI</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Quantifying the Value of AI: Going Beyond Cost Savings</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/rehgan-bleile'> <img src='https://odsc.com/wp-content/uploads/2024/10/Rehgan-Bleile.png' alt='Rehgan Bleile'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Rehgan Bleile</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Co-Founder & CEO at AlignAI | Founder at Women in Analytics (WIA)</span></div></div><div class="etn-acccordion-contents "><p> Evaluating and tracking value from AI use cases can be really challenging and ambiguous. Especially for generative AI, most methods include an estimated time savings on repeatable tasks. This talk will walk through the key techniques, like decision mapping, to leverage when initially estimated potential value and tracking or auditing value from use cases over time. We will dive into the areas that AI can and will influence decision making and how anyone can attribute outcomes and impact back to AI outputs. The audience will walk away with: 1. An understanding of decision mapping and how to apply that to AI value attribution. 2. A framework to create initial value estimations for AI use cases. 3. A way to track and measure influence from AI systems to demonstrate actual value created from AI over time.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:50 am - 12:20 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">AI for Robotics</span> <span class="firstfocus">DL</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Towards Deployable Robot Learning Systems</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/zipeng-fu'> <img src='https://odsc.com/wp-content/uploads/2024/07/Zipeng-Fu.png' alt='Zipeng Fu'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Zipeng Fu</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">PhD Student at Stanford University</span></div></div><div class="etn-acccordion-contents "><p> The field of robotics has recently witnessed a significant influx of learning-based methodologies, revolutionizing areas such as manipulation, navigation, locomotion, and drones. This talk aims to delve into the forefront of robot learning systems, particularly focusing on their scalability and deployability to open-world problems, through two main paradigms of learning-based methods for robotics: reinforcement learning and imitation learning.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>12:10 pm - 12:40 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs & RAG</span> <span class="firstfocus">AI Agents</span> <span class="secfocus">Beginner-Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">RAG in 2024: Advancing to Agents</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/laurie-voss'> <img src='https://odsc.com/wp-content/uploads/2024/06/Laurie-Voss_.png' alt='Laurie Voss'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Laurie Voss</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">VP, Developer Relations at LlamaIndex</span></div></div><div class="etn-acccordion-contents "><p> Retrieval-augmented generation is an essential tool for building modern information-retrieval systems, but it isn't enough. In this talk, we make the case that while RAG is necessary, it's not sufficient: you need to add agentic strategies to your system. We discuss the basic components of an agentic system including routing (selecting between sources), memory (providing context between queries), planning (what should we do?), reflection (did we correctly do what we intended?) and tool use. We also discuss agentic reasoning strategies including sequential (chain of thought), DAG-based, and tree based (tree of thought). Finally we dip briefly into further extensions to agents including observability, controllability and customizability.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>12:10 pm - 12:40 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Solution Showcase: Revolutionizing Data Management</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/mary-vue'> <img src='https://odsc.com/wp-content/uploads/2024/10/Mary-Vue-Syncari.png' alt='Mary Vue'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Mary Vue</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;"> VP of Partnerships and Marketing at Syncari</span></div></div><div class="etn-acccordion-contents "><p> Struggling with fragmented, inconsistent master data across systems? Syncari’s Autonomous Data Management (ADM) platform, with its integrated data control plane and cross-domain process layer, automates the unification, cleansing, and synchronization of master data and augmented data across domains like marketing, sales, and finance. Trusted by enterprises like Red Hat, CorroHealth, and FTI Consulting, Syncari ensures real-time accuracy, governance, and faster time to value. Syncari enables your data sets to be AI-ready and accessible via SQL, data exports, and APIs, allowing you to easily integrate clean, augmented data into your data science workflows for advanced analytics and modeling. Deliver rich, accurate, and fresh data for operations, customer experiences, and BI insights while streamlining process automation. Join us for the demo to see Syncari and the future of master data!</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>12:10 pm - 12:40 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="firstfocus">Generative AI</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Intro to AI Auditing - a guide for executives to navigate risk</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/cal-al-dhubaib'> <img src='https://odsc.com/wp-content/uploads/2021/02/Cal-Al-Dhubaib.png' alt='Cal Al-Dhubaib'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Cal Al-Dhubaib</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Head of AI and Data Science at Further</span></div></div><div class="etn-acccordion-contents "><p> As AI reshapes business operations, it brings unique risks that traditional audits may not fully cover. Whether you’re with a public company or a tech scale-up, boards are increasingly focused on managing these new risks. In this session, we’ll explore how to evolve internal audit practices to address specific challenges posed by AI and generative AI. You’ll learn strategies for uncovering vulnerabilities and mitigating risks in alignment with emerging regulations. And as a bonus, you’ll pick up the language and insights that can help you stand out as a trusted advisor to your board. This talk is designed for business executives and data practitioners who want a better grasp of AI risk management. We’ll cover the distinctive challenges of AI audits, including evolving standards and practical tools and credentials to help your organization prepare for effective AI oversight.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>12:10 pm - 12:40 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Data Engineering</span> <span class="secfocus">Beginner - Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Efficient Incremental Processing with Apache Iceberg and Netflix Maestro</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/jun-he'> <img src='https://odsc.com/wp-content/uploads/2024/09/Jun-He.png' alt='Jun He'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Jun He</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Staff Software Engineer at Netflix</span></div></div><div class="etn-acccordion-contents "><p> Incremental processing, an approach that processes only new or updated data in workflows, substantially reduces compute resource costs and execution time, leading to fewer potential failures and less need for manual intervention. However, enabling incremental processing on large-scale data pipelines and workflows presents significant challenges around scalability, ease of adoption, and user experience. In this talk, we will discuss how we are leveraging Apache Iceberg and Netflix Maestro to build an Incremental Processing Solution (IPS) that enables incremental processing of only new or changed data, reducing compute costs and processing times while ensuring data accuracy and freshness. By combining Iceberg's metadata capabilities for snapshots and data files with Maestro's workflow orchestration, we can efficiently handle late-arriving data and backfills in various different scenarios beyond append only mode. We will share our experiences and insights into how this IPS has empowered our data engineering teams to build more reliable, efficient, and scalable data pipelines, unlocking new data processing patterns. Through real-world use cases, we will demonstrate how IPS has significantly improved resource utilization, reduced execution times, and simplified pipeline management, all while maintaining data integrity. Additionally, we will discuss the emerging incremental processing patterns that we have discovered, such as using captured change data for row-level filtering and leveraging range parameters in business logic, as well as the techniques, best practices, and lessons learned from our journey towards incremental processing at Netflix.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>12:10 pm - 12:40 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">AI Engineering</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Powering AI with Endless Data from the Web</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/rafael-levi'> <img src='https://odsc.com/wp-content/uploads/2024/10/Rafael-Levi.png' alt='Rafael Levi'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Rafael Levi</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Senior Solutions Architecture Expert at Bright Data</span></div></div><div class="etn-acccordion-contents "><p> This talk will explore key trends in web data collection, including the shift from proxies to API-driven solutions and the growing demand for data to train AI models. We’ll showcase how Bright Data's full-stack solution—covering everything from proxies to structured datasets—enables scalable, ethical data acquisition. Attendees will learn how to streamline data collection, tackle the "long-tail" challenge of niche site scraping, and gain competitive advantage through real-time insights.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>12:35 pm - 1:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Data Engineering</span> <span class="firstfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Dimensional Data Modeling in the Modern Era</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/dustin-dorsey'> <img src='https://odsc.com/wp-content/uploads/2024/02/Dustin-Dorsey-1.png' alt='Dustin Dorsey'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Dustin Dorsey</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Principal Data Architect at Onix</span></div></div><div class="etn-acccordion-contents "><p> Dimensional data modeling has been the cornerstone of data warehousing and business intelligence for decades, but its relevance is increasingly questioned in today’s rapidly evolving data landscape. With the emergence of big data, cloud computing, and AI-driven analytics, many wonder if the traditional principles of dimensional modeling still hold value. In this session, we’ll take a deep dive into the evolution of dimensional modeling, exploring how it has adapted to modern technologies like data lakes, real-time analytics, and scalable cloud platforms. We’ll also discuss why, despite these technological advancements, dimensional modeling continues to be a vital framework for building data systems that are not only scalable and high-performing but also intuitive and easy to manage. This session will provide valuable insights for anyone looking to understand the enduring importance of dimensional modeling in the modern era.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:00 pm - 2:30 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs & RAG</span> <span class="firstfocus">Beginner - Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Data Pipeline for Retrieval Augmented Generation and Model Training at eBay</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/ramesh-periyathambi'> <img src='https://odsc.com/wp-content/uploads/2024/10/Ramesh-Periyathambi_.png' alt='Ramesh Periyathambi'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Ramesh Periyathambi</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Distinguished Engineer at eBay</span></div></div><div class="etn-acccordion-contents "><p> eBay is heavily investing in AI to drive innovation and scale its platform, supporting millions of sellers and buyers worldwide. One focus area is the use of generative AI to increase the velocity of releases and enhance developer productivity. eBay operates on thousands of code repositories across multiple programming languages and possesses a vast archive of documentation on internal frameworks, infrastructure, and features. To boost developer productivity, conversational chatbots/applications and code assistant tools powered by generative AI have been developed. The data fueling these applications are sourced from diverse origins and formats. To manage these data-intensive applications effectively, data pipelines have been engineered to extract data from sources such as GitHub, Wikis, Stack Overflow, and reformat it for use in Retrieval Augmented Generation and the training of eBay's Code LLM models. In this presentation, we will explore the overall architecture of the data pipelines driving these applications and delve into the specifics of data processing within the pipeline. Additionally, we will discuss the methods used for data chunking, the embedding model, and the vector database employed in Retrieval Augmented Generation. We will also cover the data utilized for training eBay's Code LLM, including the various model iterations. To conclude, we will summarize the evaluations conducted for both the RAG and Code LLM models.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:00 pm - 2:30 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">AI Agents</span> <span class="firstfocus">LLMs</span> <span class="secfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Building Reliable Coding Agents</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/eno-reyes'> <img src='https://odsc.com/wp-content/uploads/2024/08/Eno-Reyes.png' alt='Eno Reyes'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Eno Reyes</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">CTO at Factory</span></div></div><div class="etn-acccordion-contents "><p> Agentic system design is a rapidly evolving and intellectually fascinating field, with huge potential for transforming how software is used across industries. Unlike traditional software, agentic systems rely on non-deterministic and oftentimes difficult to predict decision making. Taking inspiration from fields like robotics, cybernetics, and biology, we can start to develop intuitions around how to build systems that are ~more~ reliable than the sum of their individual stochastic parts. Attendees will learn STOTA techniques and methods in reliable agent design.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:00 pm - 2:30 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Solution Showcase: Building an AI-Ready Workforce at Scale with DataCamp</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/yashas-roy'> <img src='https://odsc.com/wp-content/uploads/2024/10/Yashas-Roy.png' alt='Yashas Roy'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Yashas Roy</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Learning Solutions Architect at DataCamp</span></div></div><div class="etn-acccordion-contents "><p> DataCamp empowers organizations to upskill their teams in data and AI by providing an interactive and personalized approach to online learning paired with expert instruction. The platform makes it easy to learn data & AI skills at your own pace—from basic analytics to advanced AI—and gain the confidence to apply those skills on the job right away. In this session, we’ll showcase DataCamp’s approach to AI education. We will walk through DataCamp’s AI curriculum, as well as how we have incorporated AI to enhance the learning experience, and how collectively, this enables professionals at every level with the skills needed to confidently apply AI-driven solutions within their organizations. We'll also showcase DataLab, DataCamp’s AI-enabled data notebook which makes it easier to go from data to insight regardless of technical skill. DataLab leverages generative AI technology to enable you to “chat with your data”, but also comes with an IDE to review, tweak, and rerun your analysis, enabling you to seamlessly turn your work into a report that you can share with your colleagues. Whether you’re focused on developing foundational AI literacy or looking to implement cutting-edge tools in your business, this session will demonstrate how DataCamp’s platform can be a pivotal asset in your AI journey.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:00 pm - 2:30 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Generative AI</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Securing the Horizon: Safeguarding Large Language Models</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/jisheng-wang-phd'> <img src='https://odsc.com/wp-content/uploads/2024/09/Jisheng-Wang.png' alt='Jisheng Wang, PhD'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Jisheng Wang, PhD</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">VP of Engineering & Head of AI/ML at Traceable AI</span></div></div><div class="etn-acccordion-contents "><p> As Large Language Models (LLMs) become increasingly integrated into various industries, they not only usher in a new epoch of innovation but also introduce complex security challenges that could potentially undermine the technological strides being made. This talk aims to demystify the security landscape surrounding LLMs, drawing parallels with past technological advancements to outline both emergent risks and robust countermeasures. LLMs, much like their predecessors in network computing and cloud technology, have opened a new attack surface that is currently being exploited in ways that could undermine their potential. This presentation will begin with an essential overview of the Open Worldwide Application Security Project (OWASP) Top 10 Application Risks and Top 10 API Risks, explaining their relevance to LLMs. We will delve into the OWASP Top 10 LLM Specific Risks, with a focus on three primary threats: Prompt Injection, Sensitive Information Disclosure, and Training Data Poisoning. For each, real-world examples will be provided to demonstrate potential exploits, alongside effective strategies for mitigation to guide developers and executives in proactive threat management. Our objective is to elevate the understanding of LLM vulnerabilities within the AI developer community, equipping leaders with the knowledge to implement robust security measures. By fostering a security-conscious culture, we aim to keep pace with the swift advancements in AI technology. As LLMs become increasingly foundational in various applications, addressing these security challenges requires a concerted effort across multiple sectors of the security industry, including Data Security Posture Management (DSPM), Cloud Security Posture Management (CSPM), Application Security (AppSec), API Security, and Supply Chain Security. This session will not only inform but also inspire the necessary actions to secure the generative AI technologies that are becoming central to our digital existence. By the end of this talk, attendees will be equipped not only with a thorough understanding of the potential security pitfalls associated with LLMs but also with actionable insights into fortifying their AI-driven applications. The goal is to foster a proactive security culture that evolves in tandem with rapid technological advancements, ensuring that the promise of AI can be realized safely and sustainably.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:00 pm - 2:30 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Data Engineering</span> <span class="firstfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Building AI Applications with Airflow</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/steven-hillion'> <img src='https://odsc.com/wp-content/uploads/2024/10/Steven-Hillion-.png' alt='Steven Hillion'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Steven Hillion</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Head of Data and AI at Astronomer</span></div></div><div class="etn-acccordion-contents "><p> Artificial Intelligence is shaping how modern organizations make decisions, drive critical business outcomes, and support their customers and stakeholders. But behind the innovation is a continually changing set of requirements and best practices that are key to the success of getting AI into production. Join us for this webinar where resident AI/ML experts from AWS and Astronomer will explore: - Real world applications driven by Apache Airflow, including conversational AI, code generation, automated troubleshooting, and practical examples of model deployment and operational LLMs - The unique opportunities and challenges around adopting and deploying generative AI at scale - How they are also helping their collective customers drive their own AI initiatives - Top strategies to harness the potential of AI and natural language processing to drive innovation</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:00 pm - 3:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Deep Learning</span> <span class="firstfocus">Beginner - Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">New Frontiers in GenAI: From Multi-Agent Systems to On-Device LLMs and Beyond</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/dr-shelby-heinecke'> <img src='https://odsc.com/wp-content/uploads/2024/10/Shelby-Heinecke.png' alt='Dr. Shelby Heinecke'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Dr. Shelby Heinecke</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Senior AI Research Manager at Salesforce</span></div></div><div class="etn-acccordion-contents "><p> AI breakthroughs are happening at an unprecedented pace, from the development of large language models (LLMs) and large multi-modal models (LMMs), to new, emerging agentic capabilities of these models. As practitioners, it’s critical to understand the forefront of GenAI, which may power future products and new capabilities for customers. To stay at the forefront, we must look ahead at the latest AI research to understand what’s next. In this session, we will discuss emerging directions in GenAI, including multi-agent systems, which are networks of collaborative LLM-powered agents and on-device LLMs, LLMs that are small enough to fit on mobile and edge devices. We will discuss the latest techniques, challenges, and practical ways to get started.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:00 pm - 3:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="firstfocus">Beginner - Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Evaluating LLM Evaluators with a Human in the Loop</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/micaela-kaplan'> <img src='https://odsc.com/wp-content/uploads/2024/10/Micaela-Kaplan.png' alt='Micaela Kaplan'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Micaela Kaplan</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Machine Learning Evangelist at HumanSignal</span></div></div><div class="etn-acccordion-contents "><p> "Large Language Models (LLMs) are a great tool for helping us to understand how well other models or LLMs are performing, and a human-in-the-loop can be a great way to handle the more complicated or ambiguous cases. Whether we’re evaluating reference based tasks like summarization or knowledge distillation, or reference free tasks like answering open ended questions or ranking, it is important that we understand how well these LLM evaluators perform to have confidence in not only the evaluations but also in the base models themselves. This can be challenging as LLM evaluators are notoriously overconfident when providing their own confidence scores or evaluations. There are many strategies for evaluating LLM evaluators, ranging from score-based evaluations to Likert-style evaluations and more, that can help us better understand how the LLM is evaluating a particular output so that we can be confident in our models and their uses. In this talk, we’ll cover a wide range of strategies for evaluating these LLM evaluators, and explore where LLM evaluators are good, and where they might need more support by keeping a human in the loop. Using real-life examples, we’ll compare the LLM evaluations to other well known metrics for the evaluation space, allowing us to better understand what it truly means to evaluate an evaluator. We’ll also talk about the biases that may be inherent to an LLM evaluator, allowing us to more responsibly make the informed decisions that drive our businesses and problem solving. Finally, we’ll discuss how to most effectively bring a human in the loop for LLM evaluator evaluations, finding the most effective and valuable use of human knowledge and skill for helping to understand what these powerful LLM evaluators are really capable of.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:00 pm - 3:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Deep Learning</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Mind Mechanics: From Spacey States to Transformer Tech</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/swagata-ashwani'> <img src='https://odsc.com/wp-content/uploads/2024/10/Swagata-Ashwani.png' alt='Swagata Ashwani'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Swagata Ashwani</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Principal Data Scientist/Data Science Lead at Boomi</span></div></div><div class="etn-acccordion-contents "><p> Recently, there has been an growth in exploring State Space Models(SSM's) as a potential replacement for transformer models. This session aims to explore the analysis between these two powerful methodologies, delving into their practical applications and comparative performance. State Space Models have long been a staple in fields such as control theory, signal processing, and econometrics. These models excel in scenarios where system dynamics can be explicitly defined and observed over time. By representing a system's state and its evolution through transition and observation equations, SSMs provide a robust framework for time series modeling, such as forecasting economic indicators and stock market trends. The simplicity and interpretability of SSMs make them valuable for applications requiring clear, understandable results. Transformers, on the other hand, have revolutionized the way we approach sequential data, especially in natural language processing (NLP). Introduced with the groundbreaking ""Attention is All You Need"" paper, Transformers leverage self-attention mechanisms to capture long-range dependencies within data, making them highly effective for tasks like language translation, text summarization, and sentiment analysis. Models like BERT, GPT, and their successors have set new benchmarks in these domains, showcasing remarkable scalability and flexibility. In this session, we will analyze the strengths and limitations of both SSMs and Transformers through the lens of various text-based scenarios: Time Series Modeling: Compare the effectiveness of SSMs in forecasting versus the capability of Transformers in handling sequential anomalies. Language Translation: Examine how Transformers have redefined language translation tasks compared to traditional methods. Text Summarization and Sentiment Analysis: Discuss the advancements brought by Transformers in generating coherent summaries and understanding contextual sentiment. By dissecting real-world applications and performance metrics, this session will provide a comprehensive understanding of how SSMs and Transformers can be leveraged to their fullest potential. Attendees will gain insights into choosing the appropriate model based on their specific needs, ultimately guiding them towards achieving superior results in their respective domains.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:00 pm - 3:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Responsible AI</span> <span class="firstfocus">Beginner - Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Mix Integer Programming for Good, Not Just Profit</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/maria-lupetini'> <img src='https://odsc.com/wp-content/uploads/2024/10/Maria-Lupetini.png' alt='Maria Lupetini'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Maria Lupetini</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">CEO and Chief Data Scientist at InfoMaker Inc</span></div></div><div class="etn-acccordion-contents "><p> This talk introduces data science practitioners to the powerful technique of Mixed Integer Programming (MIP) and its application in solving real-world supply chain problems. Using a food bank organization as a case study, we demonstrate how MIP can optimize resource allocation and distribution decisions. I begin by outlining the challenges faced by food banks in efficiently supplying food to those in need. I then introduce MIP as a mathematical optimization method that can handle both continuous and discrete variables. I jokely called math optimzation as ""high school math on steriods"" - think Algerbra I. (Okay, middle school for all the nerdy attendees) The presentation focuses on formulating the food bank demand problem as an MIP model. I discuss key components such as objective function, constraints, and decision variables. I will show how to incorporate factors like costs, resources, and demand forecasts into the model. In my case study I show to use poverty levels in census tracts and zip codes to optimize allocation of food to areas of mostneed. I then present the solution process, using popular optimization solvers with python programming. I will interpret the results, highlighting how MIP provides actionable insights for food bank organizations to make data-driven decisions about where to increase food supply. The major food banks in San Diego county are happy with recommendations developed from these model to determine where to open the next food pantry. The talk concludes by emphasizing the broader applicability of MIP in data science, including areas like resource allocation, scheduling, and logistics. I argue that MIP deserves more attention from the in data science community due to its power in solving complex, real-world optimization problems. By the end, attendees will have gained valuable insights into an underutilized but highly effective technique for tackling multifaceted decision-making challenges in various industries.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:00 pm - 3:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">NLP</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">From NLPer to AI Engineer: The Life of a Hipster Data Scientist</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/kristy-hollingshead-phd'> <img src='https://odsc.com/wp-content/uploads/2024/10/Kristy-Hollingshead.png' alt='Kristy Hollingshead, PhD'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Kristy Hollingshead, PhD</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Senior Data Science Lead at Further</span></div></div><div class="etn-acccordion-contents "><p> From NLPer to AI Engineer: The Life of a Hipster Data Scientist explores how, despite all the buzz around GenAI LLMs, the core principles of the underlying natural language processing (NLP) and machine learning (ML) technologies haven't changed as much as it seems. While today’s tools can achieve in seconds what once took months to code by hand, have we lost something crucial in the process? In this lightning talk, I’ll touch on: How foundational NLP concepts still underpin today’s GenAI models; The trade-offs between using pre-trained models and understanding the fundamentals; Why relying too heavily on ""black box"" solutions might limit the next generation of AI Engineers; and What we risk losing when we prioritize speed and convenience over deep learning (in the classical sense). As someone with 20 years of experience in NLP, who worked on language models before they were large!, I’ll reflect on what’s truly different—and what remains the same—amid the hype. Are we still learning the skills we need for long-term success in AI?</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:25 pm - 2:55 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs & RAG</span> <span class="firstfocus">Intermediate - Advanced</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">RAG Pipelines Letting You Down? How The Fitch Group Handles High-Similarity, Frequently Updated Document Sets in Financial Services</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/pablo-vega-behar'> <img src='https://odsc.com/wp-content/uploads/2024/10/Pablo-Vega-Behar_.png' alt='Pablo Vega-Behar'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Pablo Vega-Behar</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Head of AI Implementation at The Fitch Group</span></div></div><div class="etn-acccordion-contents "><p> In financial services, where timely and precise information is critical, conventional RAG approaches can fall short when handling very large sets of similar documents, especially if the set of documents are updated frequently. This talk will describe advanced strategies that The Fitch Group’s Emerging Tech team is using to overcome these challenges and to develop RAG pipelines based on quick iterations informed by user feedback. I will discuss typical pitfalls in common RAG systems and the methods we’ve used to improve retrieval accuracy and generation relevance. Topics will include sophisticated chunking, embedding, and indexing techniques; custom models and logic for intent classification, guardrail iteration, and hybrid retrieval methods combining dense and sparse representations. I will also discuss techniques for improving application performance on very large datasets. I will cover real use cases and examples of how our approaches were implemented, and the improvements observed over previous iterations. I’ll also discuss how to design a development environment where data scientists can run experiments and get metrics using the same pipelines that machine learning engineers set up for production purposes. I will offer actionable insights for data scientists and ML engineers. Attendees will learn how to apply these advanced techniques to their own RAG systems, ensuring they can handle high-similarity documents effectively while maintaining high performance and accuracy. The intended audience are advanced scientists and engineers who are currently developing, or already have RAG systems in production. It is expected that the audience will understand the fundamentals of retrieval-augmented generation, semantic search, and guardrails for language models.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:35 pm - 3:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Generative AI</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">"Day Two" Problems: 5 Hidden Hurdles to GenAI Success and How to Overcome Them</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/lior-gavish'> <img src='https://odsc.com/wp-content/uploads/2024/10/Lior-Gavish.png' alt='Lior Gavish'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Lior Gavish</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">CTO and Co-Founder | Monte Carlo</span></div></div><div class="etn-acccordion-contents "><p> Model trained? Check. RAG pipeline built? Check. Project pilot deployed? Check. Now what? For most data teams, the barrier to entry for GenAI isn't technology - it's process, particularly as it relates to compliance, quality, and scale. In this talk, Monte Carlo co-founder and CTO Lior Gavish discusses how teams can think more strategically about rolling out GenAI projects by going beyond the initial set of technical factors to understand how streamlined operations and processes play a role in AI success. Attendees will leave with practical takeaways about how they can build more robust AI strategies and systems and a better sense of Day Two considerations that affect your entire company, including: Ideas for data teams to build and implement data cultures built around core principles of data reliability and trust Real-world examples of best-in-class data teams who are successfully running customer-facing GenAI, and what people/process/product changes they’ve made to make it happen Best practices for operationalizing detection, triage, and resolution of data incidents that could impact the reliability of GenAI products.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:35 pm - 3:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs & RAG</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">RAG on the Edge</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/amanpreet-singh'> <img src='https://odsc.com/wp-content/uploads/2024/09/Amanpreet-Singh-1.png' alt='Amanpreet Singh'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Amanpreet Singh</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">CTO and Co-Founder at Contextual AI</span></div></div><div class="etn-acccordion-contents "><p> Retrieval-Augmented Generation (RAG) has emerged as a leading solution for deploying Large Language Models (LLMs) in real-world production scenarios. Smaller LLMs are continually improving and will transform how we use AI as we begin deploying high-quality AI directly on edge devices for consumers. In this talk, I will focus on how to enable high-quality RAG applications on edge devices through two recent research developments at Contextual AI—GRIT and OLMoE. GRIT unifies the two major components of a RAG pipeline—the retriever and the generator—enabling high-throughput and low-latency applications. OLMoE offers a state-of-the-art generator with 1 billion active parameters, supporting high-performance RAG applications with minimal GPU memory requirements.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:35 pm - 3:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="firstfocus">Data Engineering</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Safeguarding App Health and Consumer Experience with Metric-aware Rollouts</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/yixin-tang'> <img src='https://odsc.com/wp-content/uploads/2024/02/Yixin-Tang-1.png' alt='Yixin Tang'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Yixin Tang</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Engineer Manager at DoorDash</span></div></div><div class="etn-acccordion-contents "><p> Coming soon!"This session introduces metric-aware rollouts, a new feature at DoorDash that ensures app health and a seamless consumer experience during product development. Metric-aware rollouts use automated checks on standardized app quality metrics to detect and address issues like app latency, errors, and crashes during the rollout of new features. By establishing clear decision rules and automating responses to performance degradation, DoorDash can safeguard app quality while maintaining rapid innovation. The system offers standardized metrics, automated analysis, and rollbacks to streamline diagnosis and response, reducing manual tracking efforts. It also introduces a budget framework for handling short-term negative impacts on app quality. High-level architecture: - Standardized metrics: Composite and sub-metrics track app performance across key actions (e.g., page load duration, action load errors, crashes) to identify issues early. We will introduce how these metrics are defined on metric platform. - Automated analysis: Tolerance thresholds are set to allow some minor degradation, but significant issues trigger alerts and a possible rollback. The alert is set base on P90 of the observed metrics. - Automated rollbacks: If thresholds are exceeded, rollouts are paused to safeguard app quality. - Approval and budget framework: Experiment teams must prepare clawback plans if rollouts cause temporary degradation below set thresholds, and leadership approval is required for severe regressions. People can learn that metric-aware rollouts represent a significant advancement in ensuring app health and a positive consumer experience by addressing risks in the product development process. The approach is scalable and automated, offering a reliable framework for managing risks across multiple metrics. While it currently focuses on app quality, the system can extend to other business guardrails, streamlining decision-making across the company.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:35 pm - 3:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Solution Showcase: Product Launch: The Answer for AI Observability</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/dr-helen-gu'> <img src='https://odsc.com/wp-content/uploads/2024/09/Helen-Gu.png' alt='Dr. Helen Gu'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Dr. Helen Gu</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Founder and CEO at InsightFinder Inc. </span></div></div><div class="etn-acccordion-contents "><p> InsightFinder is announcing their new AI Observability solution. This session, featuring Founder and CEO Helen Gu, will cover the challenges of managing enterprise-scale AI and LLM models, and includes a demonstration of the new AI Observability solution – InsightFinder AI.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:35 pm - 3:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="firstfocus">Intermediate-Advanced</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Mastering Enterprise-Grade LLM Deployment: Overcoming Production Challenges</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/jaeman-an'> <img src='https://odsc.com/wp-content/uploads/2024/09/Jaeman-An.png' alt='Jaeman An'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Jaeman An</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Co-founder & CEO at VESSL AI</span></div></div><div class="etn-acccordion-contents "><p> This session delves into the practical challenges of deploying Large Language Models (LLMs) in production, with a focus on enterprise-grade solutions. Deploying LLMs at scale introduces unique complexities, such as managing large computational resources, optimizing model performance, ensuring security, and adhering to compliance standards. We will address these challenges head-on and provide strategies to overcome them, with an emphasis on infrastructure management, latency reduction, and model reliability in production environments. Industries like healthcare, finance, and e-commerce can benefit from understanding how to safely and efficiently integrate LLMs into their existing systems. Participants will explore real-world case studies and best practices for scaling LLMs, ensuring model consistency, and handling continuous updates. Key challenges such as managing high compute costs, reducing latency, monitoring performance, and mitigating data privacy risks will be discussed. Open-source tools such as Hugging Face Inference API, LangChain, Kubernetes, and vLLM will be used to demonstrate practical solutions for running LLMs in production. By the end of the session, attendees will have actionable insights on deploying and maintaining LLMs that meet enterprise standards for performance, security, and compliance.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>3:30 pm - 4:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Solution Showcase: Scaling AI/ML with Outerbounds: How Metaflow Powers Our Open-Source Foundation</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/ville-tuulos'> <img src='https://odsc.com/wp-content/uploads/2021/10/Ville-Tuulos.png' alt='Ville Tuulos'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Ville Tuulos</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Co-founder and CEO at Outerbounds</span></div></div><div class="etn-acccordion-contents "><p> Ville Tuulos, Co-Founder and CEO of Outerbounds, will share how Metaflow, the open-source framework developed by the founders of Outerbounds at Netflix, serves as the core of Outerbounds' comprehensive MLOps platform. Metaflow simplifies the process of building and scaling machine learning applications, and in this session, Ville will showcase the latest updates to Metaflow and demonstrate how Outerbounds extends its capabilities for enterprise needs. Attendees will learn how Outerbounds enhances Metaflow’s flexibility and scalability with enterprise-grade features such as advanced infrastructure management, security, and seamless integration across environments, both on-prem and in any cloud.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>3:30 pm - 4:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Machine Learning</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Making Cloud Computing Boring Again: Lessons Learned from Deploying Billions of Python Functions</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/sarah-johnson'> <img src='https://odsc.com/wp-content/uploads/2024/09/Sarah-Johnso.png' alt='Sarah Johnson'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Sarah Johnson</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Developer Relations at Coiled</span></div></div><div class="etn-acccordion-contents "><p> We’re not breaking ground with new LLMs or generative AI. But after managing and deploying 100,000s of Dask clusters, we’ve learned some things about distributed computing in production. This talk goes through lessons learned running 1,000,000,000s of Python functions for users in critical production settings across many companies and research groups. We'll cover lessons learned, motivated by metadata from real-world workloads: - GIL vigilance is good - K8s is too heavyweight if all you want is lots of jobs - ARM is underused - Docker doesn't work well for data science folks - GPUs and spot are easier to get when you understand availability zones - The cloud isn’t as expensive as many people think Attendees will learn about distributed computing with Dask and how to avoid common pitfalls of deploying Dask in production, with a focus on cloud computing environments.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>3:30 pm - 4:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Generative AI</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Considerations for Building Enterprise-Safe AI</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/megha-jhunjhunwala'> <img src='https://odsc.com/wp-content/uploads/2024/10/Megha-Jhunjhunwala.png' alt='Megha Jhunjhunwala'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Megha Jhunjhunwala</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Senior AI Engineer at Glean</span></div></div><div class="etn-acccordion-contents "><p> Research from IDC projects that enterprise spending on generative AI solutions will reach $151.1 billion by 2027, making it vital for today’s AI engineers to understand what goes into building a secure and effective AI product tailored for enterprise use. Join Megha Jhunjhunwala, Software Engineer at Glean, as she breaks down the critical considerations engineers should be aware of when building enterprise-safe generative AI solutions. In this session, Megha will explore the fundamental differences between developing generative AI solutions for general consumers in comparison to enterprise environments. She will cover essential requirements such as ensuring data quality and scalability, maintaining contextual relevance and the importance of permission-aware AI systems to protect sensitive information and comply with enterprise-level security protocols. The discussion will also highlight the challenges and consequences of applying large language models (LLMs) to enterprise data, including the risks of missing knowledge, lack of context, hallucinations and why traditional fine-tuning methods do not suffice the enterprise environment. Megha will also educate attendees on the concept of Retrieval Augmented Generation (RAG) and its pivotal role in delivering accurate, knowledge-grounded responses. Attendees will learn how to use LLMs to augment retrieval processes, ensuring that AI systems are both effective and secure. By the end of this presentation, attendees will walk away with practical strategies for building AI solutions that not only meet the complex demands of the enterprise but also prioritize explainability, security, compliance, and reliability—all of which are key factors in the successful deployment of AI within enterprise settings.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>4:05 pm - 4:35 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Deep Learning</span> <span class="secfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Can self-supervised models make a difference in drug discovery?</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/bharath-ramsundar-phd'> <img src='https://odsc.com/wp-content/uploads/2024/07/Bharath-Ramsundar.png' alt='Bharath Ramsundar, PhD'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Bharath Ramsundar, PhD</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">CEO at Deep Forest Sciences</span></div></div><div class="etn-acccordion-contents "><p> Abstract Coming Soon!</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>4:05 pm - 4:35 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Data Engineering</span> <span class="firstfocus">Beginner - Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Extending Data Pipelines to Workflow Microservices</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/bo-lei'> <img src='https://odsc.com/wp-content/uploads/2024/10/Bo-Lei_.png' alt='Bo Lei'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Bo Lei</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Co-founder and CTO at Fleak</span></div></div><div class="etn-acccordion-contents "><p> In the evolving landscape of data engineering, the boundaries between data pipelines and workflow microservices are blurring, offering new opportunities for efficiency and scalability. This talk explores how traditional data pipelines, designed to move and process large-scale data, can be extended to function as microservices, enabling synchronous workflows that integrate with real-time business applications. We’ll dive into practical examples, discuss the key requirements for transforming pipelines into microservices, and explore advanced features such as SQL-based processing, external service integrations, and AI-enhanced workflows. Attendees will also learn about the trade-offs between lightweight, stateless systems and larger, distributed frameworks like Flink and Spark, helping them determine the best approach for their specific use cases. This session focuses on the convergence of these technologies, offering insights into how to build scalable, flexible data-driven applications without the overhead of complex infrastructure.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>4:05 pm - 4:35 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Solution Showcase: Accelerate AI Agents with FriendliAI’s GPU-optimized Service</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/soomin-chun'> <img src='https://odsc.com/wp-content/uploads/2024/10/Soomin-Chun.png' alt='Soomin Chun'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Soomin Chun</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Software Engineer at FriendliAI</span></div></div><div class="etn-acccordion-contents "><p> "AI agents are poised to revolutionize industries, but a key challenge remains: achieving response times fast enough for real-time applications. These intelligent systems, capable of perceiving and interacting with their environment to achieve goals–are increasingly being adopted in automation, robotics, and customer service. To be truly effective, AI agents must deliver high quality, low latency responses. FriendliAI streamlines the entire process of building and deploying your own AI agent, from fine-tuning LLMs for specific use cases to providing fast, scalable inference in production. In this demo talk, we’ll delve into the three solutions within Friendli Suite–dedicated endpoints, Docker containers, and serverless endpoints, as well as explain some of the tech in our own GPU inference engine that powers it all. Our Friendli Engine boasts a multitude of optimizations, such as iteration batching, smart token caching, and GPU-level optimizations. We’ll also share information about our integrations, such as with Weights & Biases, that makes managing experiments easy. Finally, we’ll look at some real-life examples with customer success stories and demonstrate how easy it is to get started with our Friendli service today. Whether you’re building your first AI agent or scaling to production, FriendliAI has you covered."</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>4:05 pm - 4:50 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Reverse Startup Pitch</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Identifying the Next Generation of AI Startups</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/ben-wilde'> <img src='https://odsc.com/wp-content/uploads/2024/10/Ben-Wilde.png' alt='Ben Wilde'> </a></div><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/cameron-turner'> <img src='https://odsc.com/wp-content/uploads/2024/10/Cameron-Turner.png' alt='Cameron Turner'> </a></div><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/anne-dwane'> <img src='https://odsc.com/wp-content/uploads/2024/10/Anne-Dwane.png' alt='Anne Dwane'> </a></div><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/igor-tabber'> <img src='https://odsc.com/wp-content/uploads/2024/10/Igor-Tabber.png' alt='Igor Tabber'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Ben Wilde</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Head of Innovation at Georgian</span></div><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Cameron Turner</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">General Partner at Oxonian Ventures</span></div><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Anne Dwane</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Co-Founder and Partner at Village Global</span></div><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Igor Tabber</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Founder and General Partner at Cortical Ventures</span></div></div><div class="etn-acccordion-contents "><p> In this panel discussion, seasoned Vexperts, investors, and visionaries will come together to shed light on the emerging trends, ecosystems, and disruptive technologies that are fostering the growth of the next generation of AI startups. Discover the secrets to recognizing innovation, and gain insights into the unique qualities that set these startups on a path to success. Be part of the conversation that’s shaping the future of AI entrepreneurship and innovation.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>4:40 pm - 5:10 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs & RAG</span> <span class="firstfocus">Generative AI</span> <span class="secfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Bitter lessons learned while building production-quality RAG systems for professional users of academic data</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/jeremy-miller'> <img src='https://odsc.com/wp-content/uploads/2024/07/Jeremy-Miller.png' alt='Jeremy Miller'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Jeremy Miller</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Product Manager, Academic AI Platform at Clarivate</span></div></div><div class="etn-acccordion-contents "><p> There is no shortage of tutorials that can help you build a Retrieval Augmented Generation (RAG) system in a few hours or less. Nevertheless, the gap between a RAG Demo and a Production-Quality RAG System remains stubbornly difficult to cross. This talk will cover the critical challenges faced and steps needed when transitioning from a demo to a production-quality RAG system for professional users of academic data, such as researchers, students, librarians, research officers, and others. We will explore the criticality of data quality and availability, making data accessible through APIs, and techniques for making data GenAI-ready. Despite the raw power of Large Language Models, data quality and availability continue to be the foundations of any production-quality machine learning system. Vector-only retrieval systems suffer from lack of connectivity and context. We will share our solutions for leveraging the strengths of vector databases while avoiding their biggest pitfalls. We will also discuss our experience leveraging data from many disparate databases to form a coherent context for a RAG system. Chat Interfaces have skyrocketed in popularity. Meanwhile, many users are accustomed to receiving useful results even from poorly specified input queries. We will discuss how to handle a range of query types, how to build query understanding, and how to help your system find the most optimal context for the user's question. Evaluating the outputs of RAG systems programmatically is fraught. Nonetheless, there are techniques that can be used to understand quality at various stages of a RAG system. Leveraging established practices in Machine Learning and Software Engineering, we will share our guidelines for understanding the quality of a RAG system's output, and how to diagnose next steps for improvement.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>4:40 pm - 5:10 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Machine Learning</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Unsupervised Machine Learning for Responsible and Robust AI Models</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/dr-helen-gu'> <img src='https://odsc.com/wp-content/uploads/2024/09/Helen-Gu.png' alt='Dr. Helen Gu'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Dr. Helen Gu</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Founder and CEO at InsightFinder Inc. </span></div></div><div class="etn-acccordion-contents "><p> As AI-powered applications become integral to our daily lives, ensuring their reliability is paramount. Any errors or downtime in AI models can impact millions of users, potentially leading to significant disruptions. However, the inherent complexity of distributed AI applications makes them prone to various challenges, such as model drift, large language model (LLM) hallucinations, and slow response times. In this talk, I will present a comprehensive set of AI Observability solutions designed to detect model drift and perform root cause analysis using unsupervised machine learning techniques. These tools are essential for maintaining the health of AI systems, allowing teams to raise advance alerts before problems affect users. Additionally, I will share real-world case studies that highlight the importance of these techniques in providing crucial insights into why problems occur and how they can be addressed effectively. Short bio: Helen Gu is a full professor in the Department of Computer Science at North Carolina State University. She is also founder and CEO for InsightFinder, a leading AIOps and AI Observability startup company. She received her PhD degree in 2004 and MS degree in 2001 from the Department of Computer Science, University of Illinois at Urbana-Champaign. She was a research staff member at IBM T. J. Watson Research Center between 2004 and 2007. She was on sabbatical at Google as a visiting scientist in 2015. She holds 10 US/International patents and published more than 90 research papers in international journals and major peer-reviewed conference proceedings. Dr. Gu is a recipient of NSF Career Award, four IBM Faculty Awards, and two Google Research Awards, 10-year best paper award from SoCC 2020, best paper awards from ICDCS 2012 and CNSM 2010.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>4:40 pm - 5:10 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Solution Showcase: Intelligence Aviator: Your Data’s New Best Friend</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/nick-carrick'> <img src='https://odsc.com/wp-content/uploads/2024/10/Nick-Carrick.png' alt='Nick Carrick'> </a></div><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/clement-wong'> <img src='https://odsc.com/wp-content/uploads/2024/10/Clement-Wong-1.png' alt='Clement Wong'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Nick Carrick</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Solutions Consultant at OpenText</span></div><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Clement Wong</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Principal Solutions Consultant at OpenText</span></div></div><div class="etn-acccordion-contents "><p> Join us for an exciting demo and technical walkthrough of OpenText Intelligence Aviator, the revolutionary tool enabling anyone to explore their data through natural language. Powered by a cutting-edge large language model (LLM), Intelligence Aviator enables seamless, natural language conversations with your business intelligence (BI) data. This generative AI-based approach makes data analysis intuitive and user-friendly, eliminating the need for specialized skills. Discover how Intelligence Aviator empowers all users to find the answers to their data questions. By democratizing data analysis, it fosters a truly data-driven culture where everyone can contribute to informed decision-making. Experience the efficiency and cost savings as Intelligence Aviator reduces bottlenecks and report backlogs, accelerating your decision-making process. Don’t miss this opportunity to see how OpenText Intelligence Aviator can become your data’s new best friend!</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>4:40 pm - 5:10 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Data Engineering</span> <span class="firstfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Disaster Recovery Options Running Apache Kafka in Kubernetes</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/geetha-anne'> <img src='https://odsc.com/wp-content/uploads/2024/10/Geetha-Anne.png' alt='Geetha Anne'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Geetha Anne</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Sr Manager, Costumer Solutions at PureStorage</span></div></div><div class="etn-acccordion-contents "><p> Deploying a disaster recovery strategy for your Apache Kafka workloads can increase availability and reliability of your mission critical applications by minimizing data loss and downtime during unexpected disasters. So architecting a perfect Kubernetes based Kafka deployment requires careful consideration of several edge case scenarios. In this session, you will learn about : * Successful disaster recovery strategies in Kafka ecosystem like Active-active, Active-passive replication, multi regional stretched clusters etc., * How some of these DR techniques evolve, when Kubernetes is the chosen deployment platform * Automation or CI/CD tools that will help achieve this.</p></div></div></div></div></div></div> <!-- end repeatable item --> <!-- start repeatable item --><div class='etn-tab ' data-id='tab6740cdee4f8a4-2'><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>9:00 am - 9:25 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="firstfocus">AI for Robotics</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading active'><p style="width: 70%;float: left;">ODSC KEYNOTE: Reinforcement Learning with Large Datasets: a Path to Resourceful Autonomous Agents</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/sergey-levine-phd'> <img src='https://odsc.com/wp-content/uploads/2024/05/Sergey-Levine.png' alt='Sergey Levine, PhD'> </a></div></div> <i class="etn-icon etn-minus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Sergey Levine, PhD</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Associate Professor, Computer Science | UC Berkeley</span></div></div><div class="etn-acccordion-contents active"><p> Recent advances in data-driven AI methods, generative models, and other techniques that can utilize large datasets have led to remarkable advances in generalization and capability. However, to create AI systems that can flexibly and resourcefully find novel solutions to new problems and obstacles they encounter in the real world, we need learning systems that can improve and adapt autonomously. Reinforcement learning offers a potential algorithmic framework to enable this. However, to make RL methods viable for real-world systems, we need to integrate the generalization capabilities that come from training on large prior datasets with the ability of RL methods to adapt on the fly. In this talk, I will describe how we can take steps toward making this possible, and discuss potential applications in robotics and other areas.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>9:30 am - 9:55 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">ODSC KEYNOTE: Lessons Learned while Building Anthropic’s LLM, Claude</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/benjamin-mann'> <img src='https://odsc.com/wp-content/uploads/2024/09/Benjamin-Mann.png' alt='Benjamin Mann'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Benjamin Mann</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Co-founder at Anthropic</span></div></div><div class="etn-acccordion-contents "><p> Generative AI has undoubtedly been the most talked about technological innovation in recent history. Anthropic debuted its public facing Claude chatbot in 2023 and has been serving customers like Slack, Jane Street, GitLab, Perplexity, and others via its Claude API. But prior to that, the company spent over two years researching AI safety to make Claude as helpful, honest and harmless as possible. In this talk, Ben Mann, co-founder at Anthropic, will outline his top lessons learned building Claude including techniques for aligning an AI system with human values and bootstrapping scalable oversight of model outputs.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>9:35 am - 10:05 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Data Engineering</span> <span class="firstfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Building and Deploying LLM applications with Apache Airflow</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/kaxil-naik'> <img src='https://odsc.com/wp-content/uploads/2022/05/Kaxil-Naik-1.png' alt='Kaxil Naik'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Kaxil Naik</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Sr. Director of Engineering at Astronomer</span></div></div><div class="etn-acccordion-contents "><p> Behind the growing interest in Generate AI and LLM-based enterprise applications lies an expanded set of requirements for data integrations and ML orchestration. Enterprises want to use proprietary data to power LLM-based applications that create new business value, but they face challenges in moving beyond experimentation. The pipelines that power these models need to run reliably at scale, bringing together data from many sources and reacting continuously to changing conditions. This talk focuses on the design patterns for using Apache Airflow to support LLM applications created using private enterprise data. We’ll go through a real-world example of what this looks like, as well as a proposal to improve Airflow and to add additional Airflow Providers to make it easier to interact with LLMs such as the ones from OpenAI (such as GPT4) and the ones on HuggingFace, while working with both structured and unstructured data. In short, this shows how these Airflow patterns enable reliable, traceable, and scalable LLM applications within the enterprise.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>10:00 am - 10:30 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="firstfocus">Beginner-Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Large Language Models as Building Blocks</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/jay-alammar'> <img src='https://odsc.com/wp-content/uploads/2023/10/Jay-Alammar.png' alt='Jay Alammar'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Jay Alammar</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Director, Engineering Fellow (NLP) at Cohere</span></div></div><div class="etn-acccordion-contents "><p> The rise of large language models is inspiring a wide variety of application ideas and experiments. For builders to become more adept at building with LLMs, it's important to gain an understanding of using LLMs as components of advanced pipelines, and not of being a text-in text-out monolith. In this talk, Jay covers a number of LLM applications covering generative use cases, as well using language models for semantic search and data exploration.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>10:15 am - 10:45 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Data Engineering</span> <span class="firstfocus">Intermediate-Advanced</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">AI-Powered ETL Pipeline Orchestration: Multi-Agent Systems in the Era of Generative AI</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/brij-kishore-pandey'> <img src='https://odsc.com/wp-content/uploads/2024/09/Brij-kishore-Pandey_.png' alt='Brij Kishore Pandey'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Brij Kishore Pandey</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Principal Engineer at ADP</span></div></div><div class="etn-acccordion-contents "><p> This presentation explores the transformative journey of ETL (Extract, Transform, Load) orchestration, tracing its evolution from simple scheduled tasks to sophisticated AI-driven systems. We begin with the era of cron jobs, highlighting their simplicity and limitations in managing data workflows. The narrative then progresses to the advent of workflow management tools, which introduced concepts like Directed Acyclic Graphs (DAGs) to handle complex dependencies. As we delve into the cloud-native phase, we examine how distributed architectures and scalable services reshaped ETL processes. The integration of AI marks a pivotal shift, introducing predictive scheduling and intelligent resource allocation. Finally, we explore the cutting-edge realm of multi-agent AI systems, where specialized AI entities collaborate to optimize entire data ecosystems. Throughout this journey, we use a hypothetical global e-commerce platform, GlobalShop, to illustrate real-world applications and challenges at each stage. The presentation includes visual aids that demonstrate the increasing complexity and capabilities of ETL orchestration tools. By tracing this evolution, we aim to provide data engineers and architects with insights into the past, present, and future of data pipeline management. The discussion culminates in a forward-looking perspective on the potential of AI-driven orchestration to handle the volume, velocity, and variety of data in modern enterprises, setting the stage for the next generation of data integration and analytics.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>10:55 am - 11:25 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Data Engineering</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Small Data Manifesto: Data infrastructure to build bigger with less</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/ryan-boyd'> <img src='https://odsc.com/wp-content/uploads/2024/02/Ryan-Boyd.png' alt='Ryan Boyd'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Ryan Boyd</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Co-founder at MotherDuck</span></div></div><div class="etn-acccordion-contents "><p> Have you been force to write a Spark job to process a few gigabytes of data? Or, worse yet, a few hundred megabytes? Unfortunately, many data engineers and analysts have had that sad task. We’ve been sold the idea that our data is BIIIGGGG and will grow incredibly HUUUUUGEE. Why? Silicon Valley geeks [like myself] told us that over the last 20 years. The reality is though, this hasn’t happened and distributed data processing is often unnecessary and wasteful with the advancement of memory density and CPU performance. Why does this matter? We’re now in a post-ZIRP world where we need to focus on doing more with less. Luckily, the technological advancements in hardware are converging with advancements in highly-efficient data engines (like DuckDB), distributed storage and bandwidth. This talk will explore the simple joys of small data, how more data != better results, how single machines are efficient and powerful and developing locally just works. Technologies discussed: DuckDB, Web Assembly (Wasm), Small Language Models (SLMs), hybrid local <-> cloud compute and more, with analysis based on industry and academic papers. Learning Objectives: Attendees will learn to re-evaluate how they're choosing and deploying data infrastructure to focus on the most appropriate technologies for their problem set. DuckDB, Web Assembly (Wasm),</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:00 am - 11:30 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Memory Tuning</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">TRACK KEYNOTE: Removing Hallucinations by 95% with Memory Tuning: A technical deep dive</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/sharon-zhou-phd'> <img src='https://odsc.com/wp-content/uploads/2024/08/Sharon-Zhou.png' alt='Sharon Zhou, PhD'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Sharon Zhou, PhD</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">CEO & Co-Founder at Lamini</span></div></div><div class="etn-acccordion-contents "><p> Hallucinations in LLMs severely limit their applicability in critical domains, confining them to shallow use cases, where hallucinating a fact incorrectly is not critical. Hallucinations remain endemic to general-purpose LLMs, even with advanced techniques like RAG and instruction-finetuning. It turns out that hallucinations are a fundamental technical problem with a concrete solution that changes the LLM objective, called Memory Tuning. In this talk, you'll learn the conceptual framework behind this breakthrough research that reduces hallucinations by up to 95% on enterprise use cases for the Fortune 500. Dive into the technical implementation details, including the innovative Mixture of Memory Experts (MoME) model that makes Memory Tuning computationally feasible at scale. Discover how this technology paves the way for more reliable, trustworthy LLM applications across diverse, high-stakes domains.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:00 am - 11:30 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Data Engineering</span> <span class="firstfocus">Intermediate-Advanced</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">An Introduction to Data Contracts</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/mark-freeman'> <img src='https://odsc.com/wp-content/uploads/2024/09/Mark-Freeman.png' alt='Mark Freeman'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Mark Freeman</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Tech Lead, GTM Engineering at Gable</span></div></div><div class="etn-acccordion-contents "><p> Managing data quality and unexpected changes is often seen as a reactive task for data teams that often turns into fire drills-- but it doesn’t have to be that way. Companies like PayPal, GoCardless, and Convoy have successfully transformed their approach to data quality by adopting “data contracts,” an innovative architectural pattern within data engineering that addresses data quality issues upstream at the source. Mark Freeman, co-author of the upcoming O’Reilly book """"Data Contracts: Developing Production-Grade Pipelines at Scale,"""" has spent the past year collaborating with organizations to implement data contracts and refine best practices. This session will provide an early preview of the book, and cover the essential components that make up the data contract architecture. This session will also guide the audience through a practical exploration of how data contracts can be tailored to fit a variety of data use cases, from startups to enterprise-level deployments. Additionally, the session will provide valuable insights through real-world use cases, showcasing how companies have successfully put data contracts into production. Mark will also share strategies for overcoming common challenges, including how data teams can effectively communicate the value of data contracts and secure buy-in from leadership for critical data quality initiatives.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:00 am - 11:30 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Generative AI</span> <span class="firstfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Preserving Privacy in LLM Training: Transforming Sensitive Data into High-quality Synthetic Data</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/dhruv-nathawani'> <img src='https://odsc.com/wp-content/uploads/2024/10/Dhruv-Nathawani.png' alt='Dhruv Nathawani'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Dhruv Nathawani</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Applied Research Scientist at Gretel</span></div></div><div class="etn-acccordion-contents "><p> To unlock the value of sensitive datasets, enterprises need to be able to operationalize it without sacrificing privacy or utility. Synthetic data offers a promising solution to do so, providing an interface to insight-rich, real-world data that is highly-valuable for training LLMs. In this session, Gretel Co-founder and CEO Ali Golshan will introduce cutting-edge techniques in differential privacy that enable the generation of synthetic text, offering a pathway to leverage data from sensitive sources — call transcripts, patient records, and product feedback — safely and effectively. He’ll discuss how differential privacy can achieve near-perfect model accuracy, within 1% of non-private models, while ensuring robust privacy protections against data memorization or replay. Using Gretel's own synthetic data platform, he will demonstrate how differential privacy mathematically guarantees that no original data can be traced or revealed — an extremely valuable tool for developers operating in highly regulated industries like finance, law and healthcare. The next frontier for AI is moving beyond general purpose models and tools, and toward specialized applications of generative AI. To do so, it’s important that enterprises can use the entirety of their corpus of data, without sacrificing data privacy or security. With most AI giants having exhausted much of what the public data domain has to offer, this next wave of innovation will rely upon private data. This session will spotlight the transformative power of differentially private synthetic data to enhance LLM training processes across sectors, empowering organizations to capture the full value of their data securely. Attendees will walk away with a deeper understanding of differential privacy, how it can enhance LLM training, and how they themselves can transform sensitive data into safe, high-quality synthetic datasets.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:35 am - 12:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="firstfocus">AI Engineering</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Gen AI in Software Development. What should you be looking for?</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/mabel-geronimo'> <img src='https://odsc.com/wp-content/uploads/2024/07/MainPhoto-Small-Mabel-Geronimo-1.png' alt='Mabel Geronimo'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Mabel Geronimo</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Senior Solutions Engineer at GitHub</span></div></div><div class="etn-acccordion-contents "><p> Join us for an insightful session led by Mabel Gerónimo, a Solutions Engineer at GitHub, as we dive into the exciting world of AI-assisted development tools. In this session, we will explore valuable lessons learned and showcase common use cases observed from customers around the globe. We will discuss critical factors to consider when evaluating AI-assisted development tools, focusing on compliance and future product roadmaps. Additionally, you will gain an introduction to Prompt Engineering, a crucial skill for optimizing AI interactions. To make this session even more engaging, it will be packed with live demos that illustrate the power and potential of AI-assisted development tools such as GitHub Copilot in real-world scenarios. Don't miss out on this opportunity to learn and experience how AI can transform your development workflow!</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:35 am - 12:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Data Engineering</span> <span class="secfocus">Beginner</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Brick-by-Brick: Exploring the Elements of Apache Kafka®</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/danica-fine'> <img src='https://odsc.com/wp-content/uploads/2024/10/Danica-Fine-1.png' alt='Danica Fine'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Danica Fine</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Staff Developer Advocate at Confluent</span></div></div><div class="etn-acccordion-contents "><p> Have you hit a brick wall when it comes to learning Apache Kafka? Do you wish that grasping Kafka were as easy, intuitive, and fun as building your favorite LEGO® set? Why shouldn’t it be!? Let’s rebuild the world of Kafka brick-by-brick starting from the basic building blocks of the technology. We’ll leave no plate unturned as we introduce events, brokers, topics, and partitions––fundamental elements that affect how data is stored inside of this powerful distributed event streaming platform. From there, explore the wider inventory of pieces in the ecosystem––APIs and tools like Kafka Streams and Kafka Connect––that you can use to migrate, stream, and transform your data. By the end of the session, you’ll know the ins and outs of the components that form the basis of Kafka, how they ‘click’ together, and what you can build with them. At that point, only one question should remain––what will YOU make with Kafka?</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:35 am - 12:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">AI Agents</span> <span class="firstfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">From Paper to Production: Implementing Gen AI Research</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/aarushi-kansal'> <img src='https://odsc.com/wp-content/uploads/2024/10/Aarushi-Kansal_.png' alt='Aarushi Kansal'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Aarushi Kansal</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">AI Engineering at AutoGPT</span></div></div><div class="etn-acccordion-contents "><p> In the new era of Gen AI, new research papers come out what feels like almost daily, covering areas such as prompting techniques, Retrieval-Augmented Generation (RAG), fine-tuning methodologies, and models. While many of these papers offer groundbreaking ideas, they often lack accompanying code or practical guidance for implementation. In this talk, I'll demystify cutting-edge research by walking through the process of implementing these papers in real-time,within 30 minutes. With a focus on how to transform academic concepts into actionable, functional AI models and also on making these solutions accessible to a broader audience such as other engineers, non technical people and other stakeholders. Using the AutoGPT platform, you'll be shown how to bridge the gap between research and real-world application. This talk is aimed at engineers who want to be able to levarge AI within their organisation, in an accessible way. You'll learn how to build on the AutoGPT platform and tecniques to bringing your work to a wider audience. Learning Objectives and Tools: OSS Tool: AutoGPT, technical skill on implementing research papers, soft skills and techniques on making AI more accessible to non technical people.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:40 am - 12:40 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">AI Engineering</span> <span class="firstfocus">Beginner - Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Unlocking the Potential of People Analytics with Data</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/christina-zhu'> <img src='https://odsc.com/wp-content/uploads/2024/06/Christina-Zhu.png' alt='Christina Zhu'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Christina Zhu</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Developer Relations & Community Manager at Visier</span></div></div><div class="etn-acccordion-contents "><p> As we navigate through the 2020s, it is evident that the workforce has become a pivotal element of business decisions throughout every industry, especially after the changes and ramifications of COVID-19. With the growing reliance on people analytics, there has been a growing need for organizations to move beyond conventional data metrics to more advanced ones powered by data science that can intelligently shape business outcomes. In this session, we will explore the nuances of analyzing diverse datasets to create a unified analytical construct that can address the ""Last Mile"" of people analytics. This critical step involves transforming isolated data silos into a cohesive, story-driven analytics mechanism that empowers decision-makers with insights into aspects like employee engagement, collaboration networks, skill inventories, and efficiency indicators. Data visualization and data analysis is key here as we move into the future of AI and data science. We'll delve into how integrated data strategies and advanced analytical models can scale operations, enhance decision-making, and navigate the shift towards a strategic, data-centric people management landscape. Our talk emphasizes practical applications of people analytics, showcasing the transformation of workforce data into actionable intelligence for addressing key organizational challenges. We'll also discuss the importance of several DataOps strategies and developing database DevOps, which are crucial for enterprise advancement.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:40 am - 12:40 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Machine Learning</span> <span class="secfocus">Beginner - Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Designing Human-Centric AI Interfaces</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/afrozy-ara'> <img src='https://odsc.com/wp-content/uploads/2024/09/Afrozy-Ara_.png' alt='Afrozy Ara'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Afrozy Ara</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Co-founder & CEO at LuminaData</span></div></div><div class="etn-acccordion-contents "><p> In the face of rapidly growing AI adoption, data teams are under immense pressure to deliver AI-enabled products that not only enhance efficiency but are also intuitive and trustworthy. This talk addresses the critical role of design in developing AI interfaces that meet these demands by emphasizing collaboration among data, business, design, technology, and legal teams. We will explore strategies for designing AI products that add real value to users' workflows, ensuring that these tools integrate seamlessly from data querying to result presentation, without causing tool sprawl. The talk will also highlight the importance of designing for various states of work—allowing users to easily shift between detailed analysis and broader overviews, thereby enhancing productivity and facilitating better decision-making. A key premise of the session is the necessity to keep the human element at the forefront of AI development. We'll discuss how transparency in AI processes, coupled with robust compliance measures like access controls and anonymization, can foster trust and confidence among users. By providing insights into how AI algorithms make decisions, users can better understand and control their AI interactions, leading to increased reliability and adoption. With this talk, Data science professionals and leaders will learn how to build AI interfaces that empower rather than overwhelm, placing humans not just in the loop, but at the helm of decision making with AI.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:40 am - 12:40 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Deep Learning</span> <span class="firstfocus">Beginner-Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Labelling Sparse Data at Scale Using Semantic Search</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/shubham-goel'> <img src='https://odsc.com/wp-content/uploads/2024/10/Shubham-Goel.png' alt='Shubham Goel'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Shubham Goel</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Senior Machine Learning Scientist at ZEFR</span></div></div><div class="etn-acccordion-contents "><p> The talk covers how to label rarely occurring samples across social media platforms at scale with the help of Vector DBs and multi-modal models such as OpenAI CLIP, and iteratively build models on top. Multi-modal models which have been trained on vast amounts of public data contain a lot of information which can be leveraged to generate embeddings for different modalities, and then be used for approximate search using a Vector DB like Vespa/Qdrant/etc. The talk would cover using these embeddings for social media images and searching for them using human-generated prompts, which will allow a person to quickly get relevant but ""rare"" occurrences of a particular query amongst vast amounts of data.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:40 am - 12:40 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="firstfocus">Beginner-Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Leveraging LLMs for Next-Generation Recommender Systems</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/arpita-vats'> <img src='https://odsc.com/wp-content/uploads/2024/10/Arpita-Vats.png' alt='Arpita Vats'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Arpita Vats</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Senior AI Engineer at Linkedin</span></div></div><div class="etn-acccordion-contents "><p> In this session, we will dive into the effects of Large Language Models (LLMs) on recommender systems, and paradigm change in the domain. We'll look at how the special powers of LLMs—such as improved language comprehension, reasoning, and contextual awareness—outperform more conventional recommendation systems. The talk will emphasize how LLMs, even with sparse user interaction data, can provide suggestions that are more precise, contextual, and tailored. We'll talk about practical uses and case studies that show how academics and business experts use LLMs to improve user experience and recommendation quality. We'll discuss issues like prompt sensitivity and sporadic misinterpretations while appreciating how LLMs can revolutionize recommender systems. This fair investigation seeks to give participants a thorough grasp of how LLMs are changing the recommendations landscape, providing insight.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:40 am - 12:40 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Machine Learning</span> <span class="secfocus">Beginner</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Mastering High-Fidelity AI Video Generation</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/ajay-jain'> <img src='https://odsc.com/wp-content/uploads/2024/09/Ajay-Jain.png' alt='Ajay Jain'> </a></div><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/paras-jain'> <img src='https://odsc.com/wp-content/uploads/2024/09/Paras-Jain.png' alt='Paras Jain'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Ajay Jain</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">CTO and Co-founder at Genmo</span></div><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Paras Jain</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">CEO and Co-founder at Genmo</span></div></div><div class="etn-acccordion-contents "><p> In this technical talk, Paras and Ajay Jain, brothers and co-founders of AI video generation research lab Genmo, will pull back the curtain on the current landscape of AI video generation technology. This talk is a unique opportunity for developers to learn about diffusion model architecture from one of its co-inventors: Ajay, co-author of the DDPM paper. Today, almost all major image and video generation models are based on Ajay's work. Ajay and his co-researchers were also the first to develop a text-to-3D model, “DreamFusion” capable of producing high-fidelity content. Paras and Ajay's session will provide an in-depth exploration of the methodologies used and challenges faced while building Genmo's newest model, launching this September. The challenges they’ll discuss solutions for include: Optimizing GPU infrastructure to improve latency speeds and scale with growing user demand Solving the ‘incoherence’ challenge of long-context video generation with prompt design and AI video workflows Paras and Ajay will also discuss methods for increasing AI-generated video quality. For example, improving frames-per-second speed above the Hollywood standard of 24 FPS, preserving motion quality and boosting photorealism. The speakers will also share samples of video generation models in action to highlight the current landscape of model capabilities and deficiencies. Paras and Ajay will conclude with a glimpse into future developments in video generation. Model builders will leave this session with a deep understanding of the technical intricacies involved with building state-of-the-art video generation models, the challenges that still need solving, and where the next generation of innovators can begin.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>12:10 pm - 12:40 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMOps</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Llamafile: Democratizing Open Source AI</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/stephen-hood'> <img src='https://odsc.com/wp-content/uploads/2024/10/Stephen-Hood-1.png' alt='Stephen Hood'> </a></div><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/justine-tunney'> <img src='https://odsc.com/wp-content/uploads/2024/10/Justine-Tunney_.png' alt='Justine Tunney'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Stephen Hood</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Open Source AI Lead at Mozilla</span></div><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Justine Tunney</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Lead Developer at Mozilla Llamafile</span></div></div><div class="etn-acccordion-contents "><p> Mozilla's Llamafile open source project democratizes access to AI not only by making open large language models easier to use, but also by making them run fast on everyday consumer CPUs. Lead developer Justine Tunney will share the insights, tricks, and hacks that she and the project community are using to deliver these performance breakthroughs. Justine will also share a sneak peek at new capabilities that are coming to Llamafile soon. Open Source AI Lead Stephen Hood will join Justine to discuss the importance of open source AI and Mozilla's vision for the impact it will have on developers and computing more broadly.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>12:10 pm - 12:40 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="firstfocus">Generative AI</span> <span class="secfocus">Advanced</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Practical Fine Tuning Strategies for Language Models and Large Language Models</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/kevin-noel'> <img src='https://odsc.com/wp-content/uploads/2024/06/Kevin-Noel.png' alt='Kevin Noel'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Kevin Noel</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">AI/ML Lead at Uzabase/Edge Japan/US</span></div></div><div class="etn-acccordion-contents "><p> Language Models (LM) and Large Language Models (LLM) have attracted significant attention in both public and businesses due to their proficiency in natural language understanding (NLU) and generation (NLG). These models have shown capabilities in recent applications such as humn natural interfaces. Despite their capabilities, applications in business verticals require a more precise control and higher accuracy to meet both business cases as well as specific organizational requirements. This presentation explains the fine-tuning mechanism of LMs and LLMs, by looking into the fundamental mechanisms behind it as well as the various trade off in real world. At start, we introduce some general machine learning concepts, such as Representation Learning and Transfer Learning, allowing us to refine the concept of fine tuning of Language Models. We continue with an overview of actual fine-tuning methods for standard LM and their different use cases. Then, we detail some implementations of Task Heads for the fine tuning process. The impact of Task Heads is also explained along with some concrete examples in real world. In a second part, we introduce the concept of Adapters, specific to Large Language Models, as well as we describe their resource efficiency The differences between Adapters and traditional Task Heads are discussed. At the end, we focus on the integration of Fine Tuning with various retrieval mechanisms (RAG) and the interplay between those 2 methods with concrete examples. This short presentation provides some overview of fine-tuning methods for LMs and LLMs, in practical production and business setting, It highlights some fundamentals and concrete use cases in production in order to be customized for specific applications. Those insights would help the audience to deepen the understanding of current and future technical progress in LM and LLM applications.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>12:10 pm - 12:40 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Generative AI</span> <span class="firstfocus">AI Engineering</span> <span class="secfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">AI for Work: How GenAI Improve the Way We Work and Collaborate</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/sam-liang-phd'> <img src='https://odsc.com/wp-content/uploads/2024/07/Sam-Liang.png' alt='Sam Liang, PhD'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Sam Liang, PhD</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Co-founder and CEO at Otter.ai</span></div></div><div class="etn-acccordion-contents "><p> Sam Liang, co-founder and CEO of Otter.ai, discusses how emerging AI technologies will revolutionize voice communication and meetings. By capturing and analyzing conversations, GenAI assistants of the future will enhance collaboration and transform how teams work, create a knowledge base that ensures transparency across organizations, and empower professionals with personalized AI avatars that can speak and make decisions on their behalf.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>12:10 pm - 12:40 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Data Engineering</span> <span class="firstfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">From Data Mess to Data Mesh - Data Management in the Age of Big Data and Gen AI</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/j%c3%b6rg-schad'> <img src='https://odsc.com/wp-content/uploads/2024/09/Joerg-Schad.png' alt='Jörg Schad'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Jörg Schad</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Head of Engineering | Nextdata</span></div></div><div class="etn-acccordion-contents "><p> With the growing adoption of Analytics, Machine Learning, and Generative AI, many enterprises struggle to productionize initial prototypes and turn them into tangible business value. The core of these challenges often lies in effective data management and accessibility. For example, even in developing a simple chatbot using a Large Language Mode, high quality data access is crucial across multiple stages, including: - Retrieval-Augmented Generation: Accessing and integrating custom knowledge sources. - Fine-tuning: Curating and managing training data. - Metrics and Audit Logs: Ensuring traceability and compliance. While these challenges—including Data Discovery, Data Quality, and Policy Management—are not new, they become increasingly important with the growing adoption, especially in regulated environments. This talk will explore how Data Mesh can address these challenges with decentralized data ownership across large organizations. By considering Data as Product, Analytics, ML, and GenAI applications can discover and access high-quality data and enable production-ready consumption. In particular, we will discuss how """"MeshRAG,"""" a multi-layer RAG architecture, allows the identification of optimal data products across a larger Data Mesh for different use cases. Join this talk if you want to learn more about how Data Mesh principles play a crucial role in Big Data and GenAI architectures.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:00 pm - 2:30 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Deep Learning</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Wearable AI in Meta: On Device ML with Neural Interface System</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/christine-long'> <img src='https://odsc.com/wp-content/uploads/2024/09/Christine-Long.png' alt='Christine Long'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Christine Long</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Software Engineering Manager (Machine Learning Engineering Team in Reality Labs) at Meta</span></div></div><div class="etn-acccordion-contents "><p> Meta developed an EMG (electromyography) wristband integrated with smart glasses at Meta Reality Labs. This talk will walk through various challenges that we encountered when developing each generation of this wristband, what is the optimal solution that we eventually implemented to bring this product to market, and what lesson we learned that can be applied to future on-device machine learning development. One of the key challenges with implementing deep neural network models on a device like the wristband lies in its limited computational resources, hardware, power and thermal constraints compared to more powerful computing platforms like laptops or servers. This necessitates the use of lightweight yet powerful models that can perform efficiently within these constraints while still delivering acceptable accuracy levels for various applications such as gesture recognition or activity detection. To address these challenges, we applied preexisting techniques but also invented a few novel techniques to optimize machine learning models specifically for on-device deployment. Additionally, a close collaboration between hardware engineers, software developers, and researchers worked together to optimize the end-to-end system for power efficiency and performance, ensuring seamless integration of on-device machine learning capabilities into the wearable products. Moreover, developers need to consider the impact of on-device training and adaptation due to varying user profiles and environmental conditions. Multiple techniques for personalization and privacy have been compared via carefully designed benchmarking systems. Together with the consideration of hardware limitation, the best approach has been employed to improve model performance across diverse user populations while adhering to strict data privacy requirements. Another crucial aspect involves devising efficient data collection and preprocessing strategies tailored for the resource-constrained nature of wearable devices like this wristband. This involves leveraging smart data sampling techniques, compressive sensing, and customized signal processing algorithms to minimize the amount of data needed for accurate model training and inference. I will also talk about challenges when we integrate this wristband with other Meta wearable devices, and how we resolved those challenges.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:00 pm - 2:30 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Data Engineering</span> <span class="firstfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Practical Data Mesh for Enterprises</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/jay-sen'> <img src='https://odsc.com/wp-content/uploads/2024/09/Jay-Sen.png' alt='Jay Sen'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Jay Sen</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Director, Data Engineering at PayPal</span></div></div><div class="etn-acccordion-contents "><p> Unlock the full potential of your enterprise data with insightful presentation on "Practical Data Mesh for Enterprises." This session explores how the data mesh paradigm transforms traditional data management by decentralizing data ownership and treating data as a product. We'll delve into the core principles of data mesh and illustrate how practical data products can solve common enterprise challenges such as scalability bottlenecks, data silos, and slow time-to-market. More importantly, how to build data products using open source tools with minimum modification of your data pipelines to achieve most of the benefits of data mesh. It also shares a case study of how PayPal leveraged this model to bring efficiency to their data organization and achieve faster business delivery. Attendees will learn about how organizations data tables can be organized into domain specific data products and how to do it using existing open source tools to build and manage data products.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:00 pm - 2:30 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Deep Learning</span> <span class="firstfocus">ML</span> <span class="secfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Neural Operators: A new era of scientific computing</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/kamyar-azizzadenesheli-phd'> <img src='https://odsc.com/wp-content/uploads/2024/05/Kamyar-Azizzadenesheli-1.png' alt='Kamyar Azizzadenesheli, PhD'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Kamyar Azizzadenesheli, PhD</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Research Staff at NVIDIA</span></div></div><div class="etn-acccordion-contents "><p> The fabric of our daily lives, from weather forecasts to stock market predictions, from the aerodynamics of vehicles to the development of innovative materials, and even in the realms of medicine and space exploration, relies heavily on scientific and engineering computing. While superintelligence in AI has made significant strides in language processing, visual recognition, and audio analysis, its potential in the vast domain of natural sciences and engineering remains largely untapped. In this talk, we delve into the evolution of AI from neural networks to neural operators, unlocking new frontiers in advanced scientific computing. Join this talk as we explore how these cutting-edge technologies are revolutionizing our approach to understanding and modeling the complexities of the natural world, paving the way for groundbreaking discoveries and innovations.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:00 pm - 2:30 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Generative AI</span> <span class="firstfocus">ML</span> <span class="secfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Scaling AI Initiatives in Retail</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/fatih-nayebi'> <img src='https://odsc.com/wp-content/uploads/2024/06/Fatih-Nayebi-1.png' alt='Fatih Nayebi'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Fatih Nayebi</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Vice President, Data & AI at ALDO Group</span></div></div><div class="etn-acccordion-contents "><p> In the dynamic landscape of retail, leveraging artificial intelligence (AI) is not just innovative but essential for maintaining competitive advantage and enhancing consumer experiences. My talk, ""Scaling AI Initiatives in Retail,"" draws on my extensive experience as Vice President of Data & AI at ALDO Group and as a faculty lecturer at McGill University, specializing in enterprise data science. This session will provide a comprehensive overview of the practical applications and strategic scaling of AI in the retail sector, focusing on revenue growth management and the transformative impact of generative AI on both employees and consumers. At ALDO, we have embarked on numerous AI-driven projects aimed at optimizing revenue growth management. By integrating AI into our pricing strategies, inventory management, and promotional activities, we have not only enhanced operational efficiencies but also maximized profitability across multiple channels. This talk will delve into specific case studies where AI applications have led to measurable improvements in sales and customer engagement. Furthermore, the advent of generative AI has revolutionized how we interact with and serve our employees and customers. By implementing generative AI tools, we have been able to automate and personalize customer interactions at scale, leading to increased customer satisfaction and loyalty. For employees, generative AI has been pivotal in creating training programs that are tailored to individual learning styles and performance metrics, thereby enhancing workforce productivity and satisfaction. Throughout the session, attendees will gain insights into the challenges and successes of implementing AI projects, from pilot to full-scale deployment. I will share best practices for building a robust data infrastructure, developing AI talent within a retail organization, and integrating AI with existing technological systems to ensure seamless adoption and scalability. The presentation aims to not only share knowledge but also to inspire action. Attendees will leave with a clear understanding of how to approach AI initiatives in their own businesses, equipped with practical strategies and examples of real-world applications in a large retail enterprise. This dialogue will also serve as a platform for fostering further innovation and collaboration among data science professionals and business leaders attending ODSC. Join me in exploring the future of AI in retail, where data-driven decisions and AI applications become the cornerstone of growth and consumer connection.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:35 pm - 3:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="firstfocus">Generative AI</span> <span class="secfocus">Intermediate-Advanced</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">How LLMs Might Help Scale World Class Healthcare to Everyone</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/vivek-natarajan'> <img src='https://odsc.com/wp-content/uploads/2024/09/Vivek-Natarajan.png' alt='Vivek Natarajan'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Vivek Natarajan</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Research Lead at Google</span></div></div><div class="etn-acccordion-contents "><p> In recent years, the field of AI has been revolutionized by the emergence of Transformers and Large Language Models. However, perhaps nowhere is their impact likely to be more profound than in the biomedicine where they have the potential to act as care multipliers, help improve our understanding of biology and solve the burden of diseases. In this talk, I will introduce recent works from my team at Google AI, Med-PaLM, Med-PaLM 2, Med-PaLM M, AMIE and Med-Gemini which I believe are key milestones towards such a future. Med-PaLM and Med-PaLM 2 were the first AI systems to obtain passing and expert level scores on US Medical License exam questions respectively, a long standing grand challenge in AI. Med-PaLM M was the first demonstration of a generalist, multimodal, biomedical AI system. More recently, two recent studies highlight AMIE's promising capabilities. In a double-blind, randomized study, AMIE performed competitively against Primary Care Physicians in text consultations. Additionally, a separate study demonstrated AMIE's significant assistive potential for clinicians facing complex diagnostic challenges. Finally, in our most recent work, we introduced Med-Gemini models which are state of the art on several medical benchmarks spanning text, images, surgical videos, EHRs, waveforms, genomics and more. I will outline the motivation, principles and technical innovations underpinning these systems. Finally, I will sketch out a vision for how we might be able to leverage such powerful systems to help scale world class healthcare to everyone and make medicine a humane endeavor again.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:35 pm - 3:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Data Visualization & Analysis</span> <span class="secfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Beyond Simple A/B Testing: Advanced Experimentation Tactics</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/dr-timothy-chan'> <img src='https://odsc.com/wp-content/uploads/2024/06/Dr.-Timothy-Chan.png' alt='Dr. Timothy Chan'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Dr. Timothy Chan</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Head of Data Science at Statsig</span></div></div><div class="etn-acccordion-contents "><p> A/B testing is rapidly establishing itself as a core tool in product development. In this talk, we will start with a recap of standard A/B testing, including best practices. We'll also explore cutting-edge, less-familiar but powerful methodologies which address well-known limitations of standard A/B Testing. These include Sequential Testing, Multi-Armed Bandits, Switchback Experiments, Stratified Sampling, Heterogeneous Effects Detection and Experimental meta-analysis. Designed for data professionals and product builders, this presentation aims to inspire the embrace of innovative approaches and provide insights into the frontiers of experimentation.Beyond Simple A/B Testing: Advanced Experimentation Tactics.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:35 pm - 3:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">AI Engineering</span> <span class="firstfocus">MLOps</span> <span class="secfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Chronon - Open Source Data Platform for AI/ML</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/varant-zanoyan'> <img src='https://odsc.com/wp-content/uploads/2024/06/Varant-Zanoyan.png' alt='Varant Zanoyan'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Varant Zanoyan</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Software Engineer at Airbnb</span></div></div><div class="etn-acccordion-contents "><p> Chronon is an open source real-time data platform for AI/ML, developed and maintained by Airbnb and Stripe. It is responsible for computing, backfilling, and serving feature transformations, with a strong emphasis on real-time computation, scalability and consistency. This talk will go over the business impact of the platform, as well as provide a technical overview of the architecture and implementation. It would be interesting to anyone who is thinking about deploying data solutions within their organization to better support AI/ML workflows. For business impact, use cases such as anti-fraud, personalization, and customer support will be explored. These use cases span both predictive ML as well as applications of generative AI solutions. Context will be given on these use cases, and why ML practitioners in these areas often hit issues when dealing with data. During the technical overview, the talk will cover details of the offline compute engine, largely orchestrated in Spark. It will also cover the online side of data computation, including the lambda architecture implemented by Chronon under the head, with implementations on both Spark Streaming and Flink. During this section, there will be an emphasis on scalable computation of complex time-windowed aggregations. It will also cover the online APIs that Chronon offers for data fetching, and illustrate the end-to-end ML flows that one can build using these components. The talk will also cover some of the details of the open source roadmap, and offer insight on how to use the project within your organization. It will leave listeners with a clear sense of what value the platform can deliver, and what next steps to take if they are interested in evaluating it.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:35 pm - 3:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Data Engineering</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Productionizing GenAI with AI Data Development</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/rebekah-westerlind'> <img src='https://odsc.com/wp-content/uploads/2024/10/Rebekah-Westerlind.png' alt='Rebekah Westerlind'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Rebekah Westerlind</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Full-stack Software Engineer | Snorkel AI</span></div></div><div class="etn-acccordion-contents "><p> We’ve all heard some form of ‘your data is your differentiator’ but in the world of off-the-shelf Generative AI models, where does your data drive the most value? All of us who have worked with AI for sometime are used to the mindset that data is most important in building a model. Now you can just grab a model pre-trained by OpenAI, Google, Hugging Face etc and start generating predictions. And these predictions can be large chunks of generated content! Where does my data actually add value in this new world? With Generative AI your unique data is just as important (if not more) than traditional AI but in different ways. Join me to learn where your data can be used and how it should be prepared, managed, and applied.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>3:10 pm - 3:40 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMOps & MLOps</span> <span class="secfocus">Beginner-Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">The Future is Fine-tuned: Training and Serving Task-specific LLMs</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/arnav-garg'> <img src='https://odsc.com/wp-content/uploads/2024/09/Arnav-Garg.png' alt='Arnav Garg'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Arnav Garg</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">ML Team Lead at Predibase</span></div></div><div class="etn-acccordion-contents "><p> Many enterprises today have a clear desire to fine-tune and serve Large Language Models (LLM's) in order to build cost-effective, task-specific models. However, fine-tuning is often out of reach for most developers or unreliable with existing tooling in the market. Fine-tuning an LLM is an intricate process that consists of orchestrating and managing infrastructure, compute, and data in a consistent way. Moreover, there are often multiple ways you can fine-tune (quantization, lora, full fine-tuning, etc), resulting in varying levels of quality. Additionally, most companies would also like to deploy and fine-tune these LLM's in their own VPC so that their sensitive data never has to leave their environment. With the open-source tools like Ludwig.ai and LoRAX - the engineers platform for building with open-source AI - teams can quickly and effectively build end-to-end workflows to customize their LLM's with their own data. With best practice defaults to get started easily and full control for deep customization, Predibase is the easiest way to fine-tune and serve LLM's in your environment. In this interactive session and demo, we'll show you how to: - Deploy & Query an off-the-shelf LLM like LLaMa-2 in your private VPC environment - Fine-tune an LLM to create a task-specific model for your use case such as code generation - Evaluate and compare performance across fine-tuned and off-the-shelf LLM's - Deploy the fine-tuned LLM to Production - Automagically manage infrastructure including serverless deployments, right-sizing compute, complex distributed training, and more</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>3:10 pm - 3:40 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">What's Next in AI</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">The Business of Open Source AI</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/kevin-van-gundy'> <img src='https://odsc.com/wp-content/uploads/2024/10/Kevin-Van-Gundy_.png' alt='Kevin Van Gundy'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Kevin Van Gundy</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">CEO at Hypermode | Prev. COO at Vercel</span></div></div><div class="etn-acccordion-contents "><p> Open-source software unequivocally improves the quality of innovation in data science and software. However, as building, training, and maintaining these tools becomes more expensive– open questions remain about how to build sustainable businesses with open source at their core. It’s easy to see why it’s good for consumers and developers that fundamental technologies are open source, but it’s not often obvious how vendors can build healthy companies around open source. In this session, Kevin Van Gundy, as someone who has helped build several IPO-scale open-source businesses, gives an inside look at the various business models, approaches, and concerns of venture-backed companies building businesses around open-source software. He’ll discuss the drivers and advantages of building open-source software as a business. He’ll also discuss the common anti-patterns that cause open-core businesses to fail.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>3:45 pm - 4:15 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Whats Next in AI</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Innovating with Multimodality and Reasoning with Mistral AI</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/avinash-sooriyarachchi'> <img src='https://odsc.com/wp-content/uploads/2024/10/Avinash-Sooriyarachchi.png' alt='Avinash Sooriyarachchi'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Avinash Sooriyarachchi</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Senior Applied AI Engineer at Mistral AI </span></div></div><div class="etn-acccordion-contents "><p> In this talk, we will dive into how Mistral AI is creating models with advanced multimodal and reasoning abilities. We will showcase Pixtral as an exemplary model of our multimodal capabilities and demonstrate our advanced reasoning and function-calling features, which are essential for agentic work. Through a live demonstration, we will illustrate how the combination of multimodality and advanced reasoning can transform how people work.</p></div></div></div></div></div></div> <!-- end repeatable item --></div></div> <!-- schedule tab end --><br /></div><div id="fragment-2-499863582" class="clearfix be-tab-content"> <!-- schedule tab start --><div class="schedule-tab-wrapper etn-tab-wrapper schedule-tab-2"><ul class='etn-nav'><li> <a href='#' class='etn-tab-a etn-active' data-id='tab6740cdeeb2fed-0'> <span class='etn-date'>29 Oct</span> <span class=etn-day>Day 1</span> </a></li><li> <a href='#' class='etn-tab-a ' data-id='tab6740cdeeb2fed-1'> <span class='etn-date'>30 Oct</span> <span class=etn-day>Day 2</span> </a></li><li> <a href='#' class='etn-tab-a ' data-id='tab6740cdeeb2fed-2'> <span class='etn-date'>31 Oct</span> <span class=etn-day>Day 3</span> </a></li></ul><div class='etn-tab-content clearfix etn-schedule-wrap'> <!-- start repeatable item --><div class='etn-tab tab-active' data-id='tab6740cdeeb2fed-0'><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>9:30 am - 10:30 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="firstfocus">Generative AI</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading active'><p style="width: 70%;float: left;">State of the art in Generative AI: From LLMs to SLMs to Large Multimodal Models to AutoPilot to AI Agents</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/chandra-khatri'> <img src='https://odsc.com/wp-content/uploads/2021/02/chandra-khatri-1.png' alt='Chandra Khatri'> </a></div></div> <i class="etn-icon etn-minus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Chandra Khatri</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">VP, Head of AI at Krutrim</span></div></div><div class="etn-acccordion-contents active"><p> Generative AI is rapidly advancing, with state-of-the-art Large Language Models (LLMs), Large Vision and Voice Models, and sophisticated Multimodal and AI Agent frameworks transforming research and applications across various domains. This talk provides an overview of the latest developments in LLMs, Speech and Language Models (SLMs), Large Multimodal Models, AutoPilot technologies, and AI Agents, highlighting current breakthroughs and future directions. We will explore how these advancements are reshaping the application landscape, driving innovation, and opening new opportunities across industries. Join us to gain insights into the cutting-edge trends and the transformative potential of Generative AI technologies.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>9:30 am - 10:30 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="secfocus">Beginner-Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">How to Make LLMs Fit Into Commodity Hardware Again: A Practical Guide</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/oliver-zeigermann'> <img src='https://odsc.com/wp-content/uploads/2021/07/oliver-zeigermann.png' alt='Oliver Zeigermann'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Oliver Zeigermann</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Machine Learning Engineer at Techniker Krankenkasse</span></div></div><div class="etn-acccordion-contents "><p> LLMs like ChatGPT are all the hype. Using them as they are or as the key part of a RAG (Retrieval-Augmented Generation) system stretches the limits of what is possible in software development today. Unfortunately, those models typically run in the cloud either because vendors just don’t want to share their models or because there simply is no hardware you could buy in large numbers to make them run in the first place. There are, however, reasons why you would want an LLM to run on machines managed by yourself: - Cost of operation - Privacy / data protection - Latency - Full control of availability and scaling In this hands-on workshop we will show different approaches on how to make powerful LLMs fit onto affordable GPUs (like a T4) or - in special cases - even make them run on CPU. We will round this up by showing you how to evaluate and compare the performance of these small LLMs. We bring all examples for you to follow along as notebooks on Google Colab. So all you need is a laptop and a browser.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>9:30 am - 10:30 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMS & RAG</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Building Retail Copilot with Retrieval-Augmented Generation (RAG) in Azure AI Studio</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/suresh-paulraj'> <img src='https://odsc.com/wp-content/uploads/2024/10/Suresh-Paulraj_.png' alt='Suresh Paulraj'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Suresh Paulraj</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Data & AI Cloud Solution Architect | Microsoft</span></div></div><div class="etn-acccordion-contents "><p> Retrieval Augmented Generation (RAG) is a technique used to build applications that integrate data from custom data sources into a prompt for a generative AI model. RAG is a commonly used pattern for developing custom copilots - chat-based applications that use a language model to interpret inputs and generate appropriate responses. In this session, we’ll use Azure AI Studio to integrate custom data into a generative AI prompt flow.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>9:30 am - 10:30 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Generative AI</span> <span class="secfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Going Small: Twice as Fast and a Third Cheaper with Model Distillation</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/ivan-lee'> <img src='https://odsc.com/wp-content/uploads/2024/09/Ivan-Lee.png' alt='Ivan Lee'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Ivan Lee</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">CEO at Datasaur, Inc</span></div></div><div class="etn-acccordion-contents "><p> In the era of large language models (LLMs), Small Language Models (SLMs) are emerging as powerful alternatives for specialized tasks. This workshop explores the potential of SLMs and demonstrates how a collection of fine-tuned SLMs can often outperform a single multi-purpose LLM for many use cases. Workshop Highlights: Model Distillation: Learn how to distill the intelligence of an open-source large language model (Llama 3.1 405B) into a smaller, more efficient model (Llama 3.1 8B). Performance Optimization: Discover techniques to maintain quality and accuracy while achieving: 3x increase in processing speed 66% reduction in operational costs Real-world Application: Using an open-source healthcare dataset, we'll demonstrate how to leverage the expertise of the 405B model to fine-tune the 8B model for specialized medical tasks. Hands-on Experience: Follow a step-by-step guide to create your own production-ready 8B model, tailored for specific use cases. By the end of this workshop, participants will have practical knowledge of SLM fine-tuning and distillation techniques, enabling them to create powerful, efficient, and domain-specific language models for their unique applications.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>10:55 am - 11:55 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">AI Engineering</span> <span class="firstfocus">Beginner-Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Creating APIs That Data Scientists Will Love with FastAPI, SQLAlchemy, and Pydantic</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/ryan-day'> <img src='https://odsc.com/wp-content/uploads/2024/07/Ryan-Day-2.png' alt='Ryan Day'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Ryan Day</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Advanced Data Scientist at Conference of State Bank Supervisors</span></div></div><div class="etn-acccordion-contents "><p> Data scientists and machine learning engineers are an important and growing API user base. But they’re not just another set of software developers – they have unique goals and tools. Ryan will share practical tips for making APIs that data scientists will love. Ryan will give a tour of the primary tasks that data scientists perform in their daily work and the tools in the Python ecosystem that they use such as Jupyter notebooks, Airflow, FastAPI, and Streamlit. Then he will discuss how those tools interact with APIs. The primary focus of the session will be demonstrating how to implement a data scientist-friendly API. The demonstrated APIs will use Python frameworks FastAPI for the API control, SQLAlchemy for data access, and Pydantic for data validation. In addition to the APIs, the session will demonstrate the value of Software Development Kits (SDKs) for data scientist usage. Attendees will learn to create APIs and SDKs using the demonstrated technologies.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:00 am - 12:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="firstfocus">LLMs</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Building High-Quality Domain-Specific Models with Mergekit: A Cost-Effective Approach using Small Language Models</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/julien-simon'> <img src='https://odsc.com/wp-content/uploads/2020/02/Julien-Simon-1.png' alt='Julien Simon'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Julien Simon</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Chief Evangelist | Arcee.ai</span></div></div><div class="etn-acccordion-contents "><p> As the demand for specialized language models grows, researchers and developers often rely on large, computationally expensive models. However, this approach can be prohibitively costly and inefficient. Fortunately, small language models (SLMs) offer a promising alternative. When combined using the open-source Mergekit library, SLMs can yield high-quality domain-specific models that rival their larger counterparts. This technical session will highlight the benefits of using SLMs as the foundation for your domain-specific models and provide attendees with a comprehensive understanding of the model merging process. Along the way, we will dive deep and demonstrate some of the most popular algorithms available in Mergekit. We'll also discuss how the merging process fits neatly into the larger workflow of language model development, situated between pre-training and alignment. Whether you're a seasoned researcher or a developer looking to stay ahead of the curve, this session will equip you with the knowledge and expertise needed to harness the power of Mergekit and create high-quality domain-specific models cost-efficiently.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>12:05 pm - 1:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Machine Learning</span> <span class="secfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Idiomatic Polars</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/matt-harrison'> <img src='https://odsc.com/wp-content/uploads/2024/01/Matt-Harrison.png' alt='Matt Harrison'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Matt Harrison</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Python & Data Science Corporate Trainer | Consultant at MetaSnake</span></div></div><div class="etn-acccordion-contents "><p> Polars can be tricky, and there is a lot of bad advice floating around. This tutorial will cut through some of the biggest issues I've seen with Pandas code after working with the library for a while and writing three books on it. We will discuss: * Proper types * Chaining * Aggregation * Debugging</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:00 pm - 3:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="firstfocus">AI Engineering</span> <span class="secfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Machine Learning with CatBoost</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/matt-harrison'> <img src='https://odsc.com/wp-content/uploads/2024/01/Matt-Harrison.png' alt='Matt Harrison'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Matt Harrison</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Python & Data Science Corporate Trainer | Consultant at MetaSnake</span></div></div><div class="etn-acccordion-contents "><p> This workshop will show how to use CatBoost. It will demonstrate model creation, model tuning, model evaluation, and model interpretation.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:10 pm - 3:10 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="firstfocus">LLMOps & MLOps</span> <span class="secfocus">Intermediate-Advanced</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Scaling Deep Learning Training with Fully Sharded Data Parallelism in PyTorch</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/shagun-sodhani'> <img src='https://odsc.com/wp-content/uploads/2020/07/Shagun-Sodhani1.png' alt='Shagun Sodhani'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Shagun Sodhani</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Tech Lead at Meta</span></div></div><div class="etn-acccordion-contents "><p> Training large-scale machine learning models requires significant computational resources. As deep learning models continue to grow in size and complexity, traditional data parallelism approaches struggle to efficiently utilize the available hardware resources. Fully Sharded Data Parallel (FSDP) addresses this limitation by distributing the training process across multiple GPUs while maintaining efficient communication between them. This allows us to train larger models and achieve better accuracy with limited hardware constraints. In this tutorial, we will cover an overview of FSDP and learn how to scale models using this technique.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>3:20 pm - 4:20 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">AI Agents</span> <span class="firstfocus">Beginner-Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Building Reliable Voice Agents with Open Source Tools</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/sara-zanzottera'> <img src='https://odsc.com/wp-content/uploads/2024/01/Sara-Zanzottera-1.png' alt='Sara Zanzottera'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Sara Zanzottera</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">AI Engineer at Kwal</span></div></div><div class="etn-acccordion-contents "><p> Large Language Models (LLMs) are great at writing. Behind a chat interface they can chat with users almost like a real human would. But are they able to _talk_ like a human? Voice Agents are LLM-powered applications that can listen to users and talk back to them with a realistic voice, handle interruptions and improvise, while sticking to the goal they're given. In this session we will learn how they’re made, which open source tools are available to build them, and we are going to see in practice how to build one. Along the way we will see what’s their industry impact today and what are the challenges of bringing a voice agent PoC to production, with some real-world stories from our own journey from a small demo to a large-scale deployment.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>4:35 pm - 5:35 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="firstfocus">Generative AI</span> <span class="secfocus">Beginner-Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Generative Finance Without LLMs: Applying Probabilistic ML</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/deepak-kanungo'> <img src='https://odsc.com/wp-content/uploads/2023/02/Deepak-Kanungo-1.png' alt='Deepak Kanungo'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Deepak Kanungo</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">CEO at Hedged Capital LLC</span></div></div><div class="etn-acccordion-contents "><p> Generative AI, and Chat GPT-4 in particular, is all the rage these days. Probabilistic machine learning (ML) is a type of generative AI that is ideally suited for finance and investing. Unlike deep neural networks, on which ChatGPT is based, probabilistic ML models are not black boxes. These models also enable you to infer causes from effects in a fairly transparent manner. This is important in heavily regulated industries, such as finance and healthcare, where you have to explain the basis of your decisions to many stakeholders. There are several reasons why probabilistic machine learning represents the next-generation ML framework and technology for finance and investing. This generative ensemble learns continually from small and noisy financial datasets while seamlessly enabling probabilistic inference, retrodiction, prediction, and counterfactual reasoning. Probabilistic ML also lets you systematically encode personal, empirical, and institutional knowledge into ML models. Unlike conventional AI, these systems are capable of warning us when their inferences and predictions are no longer useful in the current market environment. You won’t get such quantified doubts from ChatGPT’s confident hallucinations, more commonly known as fibs and lies. Whether they're based on academic theories or ML strategies, all financial models are subject to modeling errors that can be mitigated but not eliminated. Probabilistic ML systems treat uncertainties and errors of financial and investing systems as features, not bugs. And they quantify uncertainty generated from inexact inputs and outputs as probability distributions, not point estimates. This makes for realistic financial inferences and predictions that are useful for decision-making and risk management. By moving away from flawed statistical methodologies and a restrictive conventional view of probability as a limiting frequency, you’ll move toward an intuitive view of probability as logic within an axiomatic statistical framework that comprehensively and successfully quantifies uncertainty. This tutorial will introduce you the fundamental concepts, processes and technologies so that you can get started using this powerful generative AI framework.</p></div></div></div></div></div></div> <!-- end repeatable item --> <!-- start repeatable item --><div class='etn-tab ' data-id='tab6740cdeeb2fed-1'><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:00 am - 12:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Machine Learning</span> <span class="firstfocus">All levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading active'><p style="width: 70%;float: left;">QA for ML: How we can trust AI with Food Sustainability</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/serg-masis'> <img src='https://odsc.com/wp-content/uploads/2024/10/Serg-Masis_.png' alt='Serg Masis'> </a></div></div> <i class="etn-icon etn-minus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Serg Masis</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Lead Data Scientist at Syngenta | Best Selling Author of AI/ML books</span></div></div><div class="etn-acccordion-contents active"><p> Quality is everything! Quality drives adoption, making technology, products, and businesses sustainable. Competing on price is one thing, but even a free product won't succeed without sufficient quality! This session will explore the role of Quality Assurance (QA) in AI, drawing from its rich history and evolution to underscore its significance in designing systems that are not only efficient but also trustworthy. It will also connect it with another thing we must trust: FOOD! The bounty of food we harvest yearly is a delicate balance and cannot be taken for granted. Agriculture is a system that only works because of the interdependencies between various factors such as soil health, weather patterns, crop genetics, and human management. And the decisions humans make at every level. Therefore, AI models, especially those used in food sustainability, rely on data quality, algorithms, and continuous testing to ensure they deliver reliable and actionable insights. This session will explore how AI-driven Quality Assurance can be applied to agriculture to enhance food production systems, minimize waste, and promote sustainability. Through the lens of agricultural AI, we will explore practical examples of QA at work, paving the way for a future where AI can be embraced with confidence.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:00 am - 12:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMS & RAG</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Build a RAG-based Application on SEC Filing Data</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/vino-duraisamy'> <img src='https://odsc.com/wp-content/uploads/2024/10/Vino-Duraisamy.png' alt='Vino Duraisamy'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Vino Duraisamy</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Developer Advocate at Snowflake</span></div></div><div class="etn-acccordion-contents "><p> In this workshop, you’ll build a document search assistant using vector embeddings to search SEC filings. We’ll create a processing function for PDF documents using PyPDF2 and Langchain Python libraries, build a vector store, create a chat UI in Streamlit and chat (retrieval and generation) logic that will allow you to compare LLM models for cost and performance. Finally, we’ll show you how to scale pandas when working with 1TB of data with Modin.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>12:05 pm - 1:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="firstfocus">Machine Learning</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">How to Build an Interactive Front End for Your Python Data Science Models</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/mingo-sanchez'> <img src='https://odsc.com/wp-content/uploads/2024/09/Mingo-Sanchez.png' alt='Mingo Sanchez'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Mingo Sanchez</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Senior Sales Engineer at Plotly</span></div></div><div class="etn-acccordion-contents "><p> From natural language processing to spatial analysis, data science models are becoming more advanced every day. For those models to translate into actionable insights, the data needs to be visualized in an interactive, accessible fashion. That’s where front-end analytics tools like Plotly and Dash come in! Plotly is a globally known data visualization library used by millions of people worldwide. Built on top of Plotly is Dash, an open-source framework that translates individual graphs or charts into interactive Python data apps. These web-based data apps expand far beyond what is possible with a traditional “dashboard”: they are interactive, customizable, and can incorporate all types of AI/ML libraries like ChatGPT, LangChain, and TensorFlow. Additionally, data scientists can be autonomous and own the development to deployment cycle of these data apps without the need for IT or full-stack knowledge.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>12:05 pm - 1:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="firstfocus">Data Engineering</span> <span class="secfocus">Intermediate-Advanced</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Building Big Data Workflows: NiFi, Hive, Trino, & Zeppelin</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/dr-khaled-tannir'> <img src='https://odsc.com/wp-content/uploads/2024/09/Dr.-Khaled-Tannir.png' alt='Dr. Khaled Tannir'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Dr. Khaled Tannir</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Founder / Big Data Engineer & Course Lecturer at dataXper</span></div></div><div class="etn-acccordion-contents "><p> The workshop aims to provide participants with a comprehensive understanding of popular Big Data tools, focusing on building an end-to-end solution. Throughout the workshop, attendees will gain hands-on experience with tools such as Apache NiFi, Jolt, Apache Hive, Apache Trino, Apache Zeppelin, and Apache Superset. The primary objective is to create a data flow using Apache NiFi that collects Nobel Prizes data in real time from a public REST API endpoint. This data, provided in JSON format, will be processed to extract specific attributes, convert the data into Parquet format, and store it in HDFS. Participants will begin by ingesting JSON data from the REST API using Apache NiFi. The data will then undergo a Jolt transformation, which will modify the JSON structure to meet the desired output format. Once transformed, the data will be stored as a Parquet file in the Hadoop Distributed File System (HDFS), making it available for further analysis. The workshop will also cover exploratory data analysis using Trino. Participants will create a User-Managed Hive table to read the Parquet file and use Trino to perform data analysis. Special attention will be given to techniques for flattening array columns within the data, a common challenge in working with complex data structures. Finally, participants will use Apache Superset to create a dashboard that visualizes the analyzed data, offering insights into the Nobel Prizes dataset. This workshop is designed to equip participants with practical skills in Big Data processing and analysis, from data ingestion to visualization, using a suite of powerful open-source tools. By the end of the workshop, attendees will have developed a robust understanding of how to implement and manage data pipelines, analyze complex data structures, and create meaningful visualizations.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>1:15 pm - 2:15 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMOps & MLOps</span> <span class="firstfocus">Beginner - Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Introduction to Prompt Engineering and AWS Bedrock for Backend and Data Engineers</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/elad-hirsch'> <img src='https://odsc.com/wp-content/uploads/2024/10/Elad-Hirsch.png' alt='Elad Hirsch'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Elad Hirsch</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Tech Lead, CTO Office at TeraSky </span></div></div><div class="etn-acccordion-contents "><p> Join us for a hands-on session designed for backend and data engineers looking to integrate AI into their workflows. In this one-hour session, you'll get a practical introduction to Large Language Models (LLMs) and learn the essentials of crafting effective prompts through prompt engineering. We'll explore how AWS Bedrock simplifies model deployment and fine-tuning for real-world applications. The session will begin with an overview of LLMs, followed by an introduction to prompt engineering and key techniques for crafting effective prompts. Afterward, we’ll dive into AWS Bedrock, showcasing its features and capabilities, and how it enables seamless model deployment and fine-tuning in backend and data engineering workflows. The session will include a live demonstration of setting up AWS Bedrock and deploying an LLM model, followed by a Q&A to address any questions. Whether you're new to LLMs or looking to enhance your AI capabilities, this session will provide valuable insights and actionable takeaways. Agenda: Introduction to LLM and Prompt Engineering (10-15 min) Overview of Large Language Models (LLMs) Basics of Prompt Engineering Key prompt engineering techniques and best practices AWS Bedrock Overview (25 min) Introduction to AWS Bedrock Features and Capabilities Bedrock’s role in model fine-tuning, deployment, and backend/data engineering workflows AWS Bedrock Demo (10 min) Live demonstration: Setting up AWS Bedrock and deploying an LLM model</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:00 pm - 3:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="firstfocus">Generative AI</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Essential Optimal Decision-Making: Building Models and Embracing AI's Future</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/jerry-yurchisin'> <img src='https://odsc.com/wp-content/uploads/2022/09/Jerry-Yurchisin.png' alt='Jerry Yurchisin'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Jerry Yurchisin</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Data Science Strategist at Gurobi Optimization</span></div></div><div class="etn-acccordion-contents "><p> The world focuses on data, with businesses seeing it as the best opportunity to be more efficient and make smarter decisions. That’s where mathematical optimization shines—it’s a game-changer for making not just better decisions, but optimal decisions. This hands-on workshop is perfect for anyone who wants to tap into the power of optimization to make more impactful business decisions. Whether you’re just starting out or looking to deepen your skills, this session will give you the practical know-how to start using optimization models in your work. We’ll start by talking about why optimization is so important for businesses today, who uses it, and what problems they typically solve. You’ll see how it goes beyond just predicting what might happen and actually helps you make the best possible decisions in areas like pricing, supply chain, logistics, manufacturing, and finance. Next, we’ll get our hands dirty with a step-by-step walkthrough of building a basic optimization model, using a handy modeling cheat sheet to get going. You’ll pick up some practical tips and tricks that you can start using right away in real-world situations. As we move forward, we’ll dive into the exciting world where Generative AI meets optimization. You’ll learn how GenAI can help you create, improve, and refine your models, making the whole process quicker and more adaptable to your business needs with the highlights on potential pitfalls using this technology. Finally, we’ll explore some powerful open-source tools that bring machine learning and optimization together. You’ll discover how these tools can make your work smoother and help you tackle optimization problems more effectively. </p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:00 pm - 3:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Machine Learning</span> <span class="firstfocus">Intermediate - Advanced</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Active Learning & Auto Labeling with YOLO</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/dean-pleban'> <img src='https://odsc.com/wp-content/uploads/2023/01/Dean-Pleban.png' alt='Dean Pleban'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Dean Pleban</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Co-Founder & CEO at DagsHub</span></div></div><div class="etn-acccordion-contents "><p> Machine learning is transforming industries, but many data scientists are still looking for ways to use it to optimize their own workflows and build better models. While active learning is a well-known concept, few have successfully integrated it into production. In this session, you'll discover how active learning can help you train better models by building better datasets automatically. Through a hands-on live demo, I'll walk you through building an auto-labeling pipeline for computer vision with tools like Label Studio, MLflow, and YOLO. You'll leave equipped with the knowledge to implement this patter in your own work.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>3:30 pm - 4:30 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="firstfocus">Machine Learning</span> <span class="secfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Causal Graphs: Applying PyWhy to Go Beyond Explainability</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/michelle-yi'> <img src='https://odsc.com/wp-content/uploads/2023/08/Michelle-Yi.png' alt='Michelle Yi'> </a></div><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/amy-hodler'> <img src='https://odsc.com/wp-content/uploads/2023/08/Amy-Hodler.png' alt='Amy Hodler'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Michelle Yi</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Board Member at Women In Data</span></div><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Amy Hodler</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Founder, Consultant at GraphGeeks.org</span></div></div><div class="etn-acccordion-contents "><p> In a world obsessed with making predictions and generative AI, we often overlook the crucial task of making sense of these predictions and understanding results. We can't trust our decisions and policies if we don't know how and why recommendations are made. This session looks at using the PyWhy open-source ecosystem for causal machine learning and better decision-making. In the realm of predictions, explainability, and causality, graphs have emerged as a potent model, leading to significant breakthroughs. These purposefully designed graphs capture and represent the intricate connections between entities, providing a comprehensive framework for understanding complex systems. Today, leading teams leverage this framework to surface directional patterns, compute complex logic, and as a foundation for causal inference. This training will empower you by examining how to create and incorporate causal graphs into your predictive workflow to improve solutions. You'll gain a deep understanding of foundational concepts such as Jedeau Pearl's ""do"" operator, causal discovery, and how to keep domain expertise in the loop. We'll delve into a practical example using PyWhy to evaluate city data and identify interventions that impact community resilience. We'll also explore using Causal Learn, LLMs, and other tools that expedite the complex process of modeling a problem as a causal graph. Join us as we examine graphs' transformative potential and profound impact on predictive modeling, explainability, and causality in the era of generative AI. This is an exciting time for our field, and we're thrilled to share our insights with you.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>3:30 pm - 4:30 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Machine Learning</span> <span class="firstfocus">Beginner-Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Solving Time Series Problems</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/john-mount-phd'> <img src='https://odsc.com/wp-content/uploads/2024/10/John-Mount.png' alt='John Mount, PhD'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>John Mount, PhD</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Principal Consultant at Win Vector LLC</span></div></div><div class="etn-acccordion-contents "><p> Time series methods solve the problem of forecasting a plausible future from past observations and additional external regressors. It is a bit of a neglected topic in data science and machine learning, though well understood in engineering and statistical circles. Time series forecasting presents a technical challenge due to the usual ""prediction is hard, especially about the future"" issue and the increased modeling risks due to co-linearity of variables, non-identifiability of dynamic systems, and the need to extrapolate (instead of merely memorizing and interpolating). Time series forecasting also presents operational challenges as the solutions usually bind fitting and forecasting much closer together than is typical for production models. However, time series forecasting is very valuable: as having a good estimate of the future usually has huge benefits for planning and other business activities. In this talk we will show how to avoid the pitfalls and effectively solve time series problems using open source packages in Python. We will also show which pitfalls are dangerous, and which are merely how things are usually taught. The participant should come away with ideas how to apply time series solutions to valuable opportunities of their choice. In this talk we will demonstrate Python, Stan, Prophet, and scikit-learn.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>4:35 pm - 5:35 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Building Multiple Natural Language Processing Models to Work In Concert Together</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/david-vonthenen'> <img src='https://odsc.com/wp-content/uploads/2024/10/David-vonThenen.png' alt='David vonThenen'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>David vonThenen</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Software Engineer at Deepgram</span></div></div><div class="etn-acccordion-contents "><p> 1.5 billion messages are sent in Slack every week. At Zoom's peak, 300 million virtual meetings occurred on their platform daily. Facebook hosts 260 million conversations on any given day. The amount of information and data exchanged on platforms like Facebook, TikTok, and ChatGPT is almost incomprehensible. These conversations are transforming social networks into conversation data brokers used to identify trends, associations, and changes in the world. To collect this data, we must first build Natural Language Processing (NLP) models to break down these conversations and classify what's being said to understand their context. This session will focus on creating and collecting datasets, using those datasets to develop machine learning models, and then covering strategies for leveraging multiple machine learning models for data mining. We will cover how to obtain and process conversation data from multiple audio and video input sources and how to use the NLP models created in this session to extract information or metadata (e.g., sentence classification, entity recognition, etc.). During this talk, we will have live demos and provide code/resources for everything covered in this session.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>4:35 pm - 5:35 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Machine Learning</span> <span class="secfocus">Intermediate - Advanced</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Raw Tabular Data Lakes to ML Ready Features Using Graphs</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/wes-madrigal'> <img src='https://odsc.com/wp-content/uploads/2024/10/Wes-Madrigal.png' alt='Wes Madrigal'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Wes Madrigal</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">CEO / Co-Founder at Kurve, Inc.</span></div></div><div class="etn-acccordion-contents "><p> Despite all of the progress in text and video with generative AI operationalizing ML/AI on tabular data remains a challenge. No machine learning method is capable of learning directly on data spread across multiple relational tables to date. In this talk we present methods to infer relational structure between tables in a data lake, automate feature engineering computations, and automate the creation of AI-ready datasets for any target variable on production scale data lakes and data warehouses.</p></div></div></div></div></div></div> <!-- end repeatable item --> <!-- start repeatable item --><div class='etn-tab ' data-id='tab6740cdeeb2fed-2'><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>9:40 am - 10:40 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs & RAG</span> <span class="secfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading active'><p style="width: 70%;float: left;">Building an Agentic Rag Application with LangGraph</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/valentina-alto'> <img src='https://odsc.com/wp-content/uploads/2024/08/Valentina-Alto.png' alt='Valentina Alto'> </a></div></div> <i class="etn-icon etn-minus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Valentina Alto</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">AI and Apps Tech Architect at Microsoft</span></div></div><div class="etn-acccordion-contents active"><p> In this hands-on workshop, participants will dive into the world of Agent-based Retrieval Augmented Generation (RAG), a cutting-edge approach that integrates agents with retrieval systems to build smarter, more adaptive GenAI solutions. We'll explore how combining the power of VectorDBs and LLMs with LangGraph creates dynamic workflows that enhance the retrieval and generation process. By the end of the session, participants will gain practical insights into creating multi-agent systems that utilize specialized RAG pipelines for different tasks, leveraging AI agents to make intelligent decisions.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>10:50 am - 11:50 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="firstfocus">Beginner - Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Building LLM Applications Using Open-source Models</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/sudip-shrestha-phd'> <img src='https://odsc.com/wp-content/uploads/2024/09/Sudip-Shrestha.png' alt='Sudip Shrestha, PhD'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Sudip Shrestha, PhD</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Data Science Lead/ Sr. Manager at Asi Government</span></div></div><div class="etn-acccordion-contents "><p> This session is designed for beginners interested in learning how to build applications using open-source large language models (LLMs). We will start by explaining the basics of LLMs and how they work, with a focus on popular open-source models. Attendees will be introduced to the Hugging Face ecosystem, a powerful platform that simplifies access to these models. The session will cover a code walkthrough on building LLM application using opens-source model. By the end of the talk, participants will have a clear understanding of how to get started with LLMs using open-source tools, regardless of their prior experience in AI.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:00 am - 12:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="firstfocus">Generative AI</span> <span class="secfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Function Calling with Llama-3 Running at 1,000 tokens/s</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/varun-badrinath-krishna'> <img src='https://odsc.com/wp-content/uploads/2024/08/Varun-Badrinath-Krishna-1.png' alt='Varun Badrinath Krishna'> </a></div><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/petro-junior-milan'> <img src='https://odsc.com/wp-content/uploads/2024/08/Petro-Junior-Milan.png' alt='Petro Junior Milan'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Varun Badrinath Krishna</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Sr Principal AI Solutions Engineer at SambaNova Systems</span></div><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Petro Junior Milan</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Principal AI Engineer at SambaNova Systems</span></div></div><div class="etn-acccordion-contents "><p> In this session, you will learn about function calling in open source large language models (LLMs). Function calling extends the capabilities of LLMs beyond text generation, allowing them to interact with external systems, APIs, and tools. This technique involves the model selecting appropriate functions or tools from a predefined set and invoking them through structured responses, typically formatted as JSON objects, to accomplish specific tasks. Key benefits of function calling include real-time data access, math problem solving, improved decision-making, and task execution. We will begin the session with the fundamentals of function calling, followed by a basic hands-on example, and conclude with a practical use case. We will leverage open source frameworks and libraries such as LangChain and Pydantic, and make fast LLM inferences using the open source Llama-3-8b model that runs at 1,000 tokens/s on the SambaNova AI platform powered by the SN40L AI chip. Speed is crucial in this context as it enables real-time responsiveness, significantly enhancing user interaction and experience. By completing this workshop, you will develop an understanding of function calls with LLMs. The knowledge you gain will be helpful in building advanced AI agents, as well as enhancing your generative AI workflows.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>12:00 pm - 1:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">AI Engineering</span> <span class="secfocus">Beginner-Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Using APIs in Data Science Without Breaking Anything</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/ryan-day'> <img src='https://odsc.com/wp-content/uploads/2024/07/Ryan-Day-2.png' alt='Ryan Day'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Ryan Day</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Advanced Data Scientist at Conference of State Bank Supervisors</span></div></div><div class="etn-acccordion-contents "><p> Data scientists use APis in a variety of ways: as data sources for their data pipelines, to gather data for analytics products, and to consume machine learning models. The benefit of APIs is that they’re easy to use. The downside is that they’re also easy to misuse. How can data scientists design their systems to use APIs in a fault-tolerant way that can intelligently react to errors and ensure the validity of data that is provided? How can data scientists use APIs responsibly, without bringing down the API by accident? In this hands-on workshop, attendees will learn step-by-step the process of consuming a REST API in a Jupyter notebook. They will create fault-tolerant code that validates the output of APIs and handles errors intelligently. They will learn advanced techniques such as progressive backoff to avoid breaking the system they’re trying to call. Then they will learn how to make that code reusable by creating a software development kit (SDK) for the API.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>12:05 pm - 1:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="etn-schedule-location"> <span class="firstfocus">Generative AI</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">The Developers Playbook for Large Language Model Security</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/steve-wilson'> <img src='https://odsc.com/wp-content/uploads/2024/06/SteveWilson.png' alt='Steve Wilson'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Steve Wilson</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Chief Product Officer at Exabeam</span></div></div><div class="etn-acccordion-contents "><p> As Gen AI technologies rapidly advance, the potential risks and vulnerabilities associated with Large Language Models (LLMs) become increasingly significant. This talk, based on insights from ""The Developer's Playbook for Large Language Model Security,"" published by O'Reilly Media, provides a comprehensive framework for securing LLM applications. Attendees will gain a deep understanding of common vulnerabilities, such as prompt injection, training data poisoning, model theft, and overreliance on LLM outputs. The session will explore real-world case studies and actionable best practices, illustrating how LLM applications can be safeguarded against these threats. Through examples of past security incidents, both from real-world implementations and speculative scenarios from popular culture, participants will see the potential consequences of unaddressed vulnerabilities. The talk will also cover the implementation of the RAISE framework, which stands for Responsible AI Security Engineering, designed to provide a step-by-step approach to building secure and resilient AI systems. Attendees will learn about zero trust architectures, supply chain security, and continuous monitoring practices essential for maintaining the integrity of LLM applications. The session will highlight the importance of ethical considerations in AI development, ensuring that technological advancements benefit society while minimizing risks. By the end of this talk, developers and security professionals will be equipped with the knowledge and tools needed to build, deploy, and maintain secure LLM applications, paving the way for a safer AI-driven future.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>1:00 pm - 2:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Deep Learning</span> <span class="firstfocus">Beginner-Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Topological Deep Learning Made Easy with TopoX</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/dr-mustafa-hajij'> <img src='https://odsc.com/wp-content/uploads/2024/10/Dr.-Mustafa-Hajij.png' alt='Dr. Mustafa Hajij'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Dr. Mustafa Hajij</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Assistant Professor at University of San Francisco</span></div></div><div class="etn-acccordion-contents "><p> In this tutorial, we introduce **TopoX**, a Python software suite designed for topological deep learning, enabling computation and learning on complex topological structures beyond graphs. These include hypergraphs, simplicial complexes, cellular complexes, path complexes, and combinatorial complexes. The tutorial covers the three core components of TopoX: **TopoNetX**, which provides an easy-to-use framework for constructing and performing computations on these topological domains, supporting nodes, edges, and higher-order cells; **TopoEmbedX**, which offers techniques for embedding topological domains into vector spaces, akin to graph embedding methods; and **TopoModelX**, a PyTorch-based library offering powerful higher-order message-passing functions for building neural networks specifically tailored to topological domains. Through hands-on examples and practical demonstrations, participants will learn how to leverage these tools for deep learning tasks on complex topological structures. By the end of the tutorial, attendees will be well-equipped to apply **TopoX** in their own research or projects related to topological deep learning. The suite is open-source, extensively documented, unit-tested, and available under the MIT license 🌐 TopoX 🍩 — TopoX documentation (pyt-team.github.io).</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>1:10 pm - 3:10 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs & RAG</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Building Agentic and Multi-Agent Systems with LangGraph</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/chris-alexiuk'> <img src='https://odsc.com/wp-content/uploads/2024/10/Chris-Alexiuk_.png' alt='Chris Alexiuk'> </a></div><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/greg-loughnane'> <img src='https://odsc.com/wp-content/uploads/2024/10/Greg-Loughnane.png' alt='Greg Loughnane'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Chris Alexiuk</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Head of LLMs at AI Makerspace | Founding Machine Learning Engineer at Ox</span></div><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Greg Loughnane</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Co-Founder & CEO at AI Makerspace</span></div></div><div class="etn-acccordion-contents "><p> People and companies in 2024 aim to build ever more complex and performant LLM applications. Leveraging context (e.g., Retrieval Augmented Generation, or RAG), reasoning, and access to external tools or functions (Agents), are front and center. For applications to leverage context well, they must provide useful input to the context window (e.g., [in-context learning](https://openai.com/index/language-models-are-few-shot-learners/)), through direct prompting (Prompt Engineering) or search and retrieval (RAG). To leverage reasoning is to leverage the Reasoning-Action ([ReAct](https://arxiv.org/abs/2210.03629)) pattern, and to be “agentic” or “agent-like.” Another way to think about agents is that they enhance search and retrieval through the intelligent use of tools or services. The best practice orchestration tooling in the industry for building context-aware, reasoning applications with LLMs is LangChain. To build with Agents, LangGraph leverages a graph-based approach to add cyclical reasoning loops to our application logic. Once we introduce agentic cycles, adding agents (e.g., multi-agents) to our graph becomes trivial. In this session, we'll break down all the concepts and code you need to understand and build the industry-standard agentic and multi-agent systems, from soup to nuts, with LangChain.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:00 pm - 3:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMOps & MLOps</span> <span class="firstfocus">Intermediate</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Customizing AI with Synthetic Data: Techniques and Real-World Applications</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/terry-rodriguez'> <img src='https://odsc.com/wp-content/uploads/2024/09/Terry-Rodriguez.png' alt='Terry Rodriguez'> </a></div><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/salma-mayorquin'> <img src='https://odsc.com/wp-content/uploads/2024/09/Salma-Mayorquin.png' alt='Salma Mayorquin'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Terry Rodriguez</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Co-Founder at Remyx AI</span></div><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Salma Mayorquin</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">CEO at Remyx AI</span></div></div><div class="etn-acccordion-contents "><p> Recent research suggests that utilizing synthetic data and enhancing data curation when refining models can result in significant performance enhancements. This tutorial delves into the methods for creating, curating, and processing synthetic data to tailor and enhance AI systems for various use cases. We present three detailed use cases: enhancing RAG-based systems, training LLMs for tool use, and customizing chatbots, including the techniques, methodologies, and code. The session aims to equip participants with the necessary techniques, methodologies, and tools to effectively use synthetic data, identify the most effective data curation strategies for their specific applications, and implement automated and scalable evaluation systems to enhance overall application performance.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>3:05 pm - 4:05 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="secfocus">Beginner</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">On-Device Multimodal Model development and Inference Acceleration</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/zack-li'> <img src='https://odsc.com/wp-content/uploads/2024/09/Zack-Li.png' alt='Zack Li'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Zack Li</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">CTO at Nexa AI</span></div></div><div class="etn-acccordion-contents "><p> In large language models, traditional tokenization methods have been effective for handling language-related tasks, but they face significant challenges when applied to function calling, often resulting in inaccuracies and hallucinations. To overcome this, we have introduced a novel approach in the Octopus model by using functional tokens, transforming function calling into a language completion task. By treating functions as distinct tokens, the Octopus model enhances accuracy and reliability in function execution, allowing seamless integration of function calling capabilities into language models. Additionally, the challenge of running AI models efficiently on-device, especially with respect to power consumption and processing speed, remains a key hurdle. Our solution addresses this by incorporating advanced quantization and optimization techniques that make on-device AI both faster and more energy-efficient. In this session, we will discuss general quantization technology and local model inference for various modalities such as image generation, text generation, image understanding, and automatic speech recognition. Attendees will learn how these innovations can improve function calling accuracy and significantly boost the efficiency of on-device AI applications, making cutting-edge AI accessible across a range of industries and devices.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">AI Engineering</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Efficient Data Pipelines for AI</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/this-session-will-feature-a-leading-expert-in-the-field'> <img src='https://odsc.com/wp-content/uploads/2024/07/Screenshot-2024-07-22-at-19.03.02-Cropped.png' alt='This session will feature a leading expert in the field'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>This session will feature a leading expert in the field</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Check back for more information</span></div></div><div class="etn-acccordion-contents "><p> Building robust and efficient data pipelines is critical for the success of any AI project. This session explores strategies for designing, implementing, and optimizing data pipelines to support AI workloads. We will delve into topics such as data ingestion, preprocessing, transformation, and feature engineering. Best practices for handling large-scale datasets, ensuring data quality, and accelerating pipeline performance will be discussed. Attendees will gain practical insights into building data pipelines that can effectively feed AI models and drive business value.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">AI Agents</span> <span class="etn-schedule-location"> <span class="firstfocus">LLMs&RAG</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">LLM and Agent Chaining</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/this-session-will-feature-a-leading-expert-in-the-field'> <img src='https://odsc.com/wp-content/uploads/2024/07/Screenshot-2024-07-22-at-19.03.34-Cropped.png' alt='This session will feature a leading expert in the field'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>This session will feature a leading expert in the field</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Check back for more information</span></div></div><div class="etn-acccordion-contents "><p> LLM and agent chaining is emerging as a powerful paradigm for creating complex AI systems that can reason, plan, and execute tasks in a multi-step fashion. This session explores the concept, benefits, and challenges of chaining LLMs and agents together. We will delve into techniques for breaking down complex problems into subtasks, managing state and context, and ensuring effective collaboration between LLMs and agents. Attendees will gain insights into building intelligent systems capable of handling intricate and dynamic environments.</p></div></div></div></div></div></div> <!-- end repeatable item --></div></div> <!-- schedule tab end --><br /></div><div id="fragment-3-499863582" class="clearfix be-tab-content"> <!-- schedule tab start --><div class="schedule-tab-wrapper etn-tab-wrapper schedule-tab-2"><ul class='etn-nav'><li> <a href='#' class='etn-tab-a etn-active' data-id='tab6740cdeeccc42-0'> <span class='etn-date'>28 Oct</span> <span class=etn-day>Day 0</span> </a></li><li> <a href='#' class='etn-tab-a ' data-id='tab6740cdeeccc42-1'> <span class='etn-date'>29 Oct</span> <span class=etn-day>Day 1</span> </a></li><li> <a href='#' class='etn-tab-a ' data-id='tab6740cdeeccc42-2'> <span class='etn-date'>30 Oct</span> <span class=etn-day>Day 2</span> </a></li></ul><div class='etn-tab-content clearfix etn-schedule-wrap'> <!-- start repeatable item --><div class='etn-tab tab-active' data-id='tab6740cdeeccc42-0'><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>10:00 am - 12:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Data Visualization & Analysis</span> <span class="etn-schedule-location"> <span class="firstfocus">Machine Learning</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading active'><p style="width: 70%;float: left;">A Practical Introduction to Data Visualization for Data Scientists</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/robert-kosara'> <img src='https://odsc.com/wp-content/uploads/2024/02/Rober-Kosara_.png' alt='Robert Kosara'> </a></div></div> <i class="etn-icon etn-minus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Robert Kosara</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Data Visualization Developer at Observable</span></div></div><div class="etn-acccordion-contents active"><p> How does data visualization work, and what can it do for you? In this workshop, data visualization researcher and developer Robert Kosara will teach you the basics of how and why to visualize data, and show you how to create interactive charts using open-source tools. You'll learn… - the fundamental building blocks of data visualization: visual variables, data mappings, etc. - the difference between continuous and categorical data, and what it means for data visualization and the use of color - what grammars of graphics are (the 'gg' in 'ggplot'!) and how they help make more interesting visualizations - the basic chart types, how they work, and what they're best used for - a few unusual chart types and when to use them - how to prepare data for common data visualization tools - how to build a simple interactive modeling tool that combines observed and modeled data in a single visualization - when to use common charts vs. when to go for bespoke or unusual visualizations We'll build all these visualizations using the open-source Observable Plot framework, but the concepts apply similarly to many others (such as ggplot, vega-lite, etc.). To follow along, you'll need a computer with an editor (such as Visual Studio Code) as well as a download of the project we provide (see the prerequisites).</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>10:00 am - 12:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Statistics</span> <span class="firstfocus">Beginner</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Statistics and Hypothesis Testing</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/thomas-nield'> <img src='https://odsc.com/wp-content/uploads/2023/08/Thomas-Nield.png' alt='Thomas Nield'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Thomas Nield</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Instructor at University of Southern California | Founder at Nield Consulting Group and Yawman Flight</span></div></div><div class="etn-acccordion-contents "><p> Statistics and hypothesis testing are the foundation of all our data-driven innovations including machine learning and generative AI. But with all this availability of data and modeling, it is easy to lose sight of the scientific method and its role. In this session, we will learn the fundamentals of descriptive and inferential statistics, and how they relate to machine learning and data mining. This will include understanding the relationship between a sample and a population, the p-value, and how we measure truth. We will also talk about the dangers of p-hacking and how it arises in data-driven environments.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>12:30 pm - 2:30 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Machine Learning</span> <span class="secfocus">Beginner</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Introduction to Machine Learning with Python</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/sudip-shrestha-phd'> <img src='https://odsc.com/wp-content/uploads/2024/09/Sudip-Shrestha.png' alt='Sudip Shrestha, PhD'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Sudip Shrestha, PhD</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Data Science Lead/ Sr. Manager at Asi Government</span></div></div><div class="etn-acccordion-contents "><p> The """"Introduction to Machine Learning with Python"""" is designed for those seeking to understand the growing field of Machine Learning (ML), a key driver in today’s data-centric world. This training offers foundational knowledge in ML, emphasizing its importance in various industries for informed decision-making and technological advancements. Participants will learn about different ML types, including supervised and unsupervised learning, and explore the complete lifecycle of an ML model—from data preprocessing to deployment. The course highlights Python’s role in ML, introducing essential tools and libraries for algorithm implementation. A practical component involves hands-on implementation of an ML use case, consolidating theoretical knowledge with real-world application. Ideal for beginners, this course provides a comprehensive yet concise introduction to ML, equipping attendees with the skills to apply ML concepts effectively in diverse scenarios.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>12:30 pm - 2:30 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs & RAG</span> <span class="firstfocus">Beginner</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Introduction to Retrieval-Augmented Generation (RAG)</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/sheamus-mcgovern'> <img src='https://odsc.com/wp-content/uploads/2023/10/Sheamus-McGovern.png' alt='Sheamus McGovern'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Sheamus McGovern</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">CEO and Software Architect, Data Engineer, and AI expert at ODSC</span></div></div><div class="etn-acccordion-contents "><p> In this 2-hour hands-on training, participants will explore the fundamentals of Retrieval-Augmented Generation (RAG), a framework that combines the power of retrieval-based methods with generative models to deliver more accurate and contextually relevant outputs. The session will cover essential stages such as data loading, indexing, querying, and generating responses using large language models (LLMs). Through practical exercises in Python, attendees will implement a basic RAG pipeline using popular frameworks like LangChain and LlamaIndex, and apply these techniques in real-world scenarios like question-answering and content generation. By the end of the session, participants will have a solid understanding of how to integrate retrieval mechanisms into generative AI workflows, and the skills to build robust RAG systems.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Machine Learning</span> <span class="secfocus">Self Paced</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Introduction to R</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/odsc-instructor'> <img src='https://odsc.com/wp-content/uploads/2024/01/blue.png' alt='ODSC Instructor'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>ODSC Instructor</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;"></span></div></div><div class="etn-acccordion-contents "><p> Dive into the world of R programming in this interactive workshop, designed to hone your data analysis and visualization skills. Begin with a walkthrough of the Colab interface, understanding cell manipulation and library utilization. Explore core R data structures like vectors, lists, and data frames, and learn data wrangling techniques to manipulate and analyze datasets. Grasp the basics of programming with iterations and function applications, transitioning into Exploratory Data Analysis (EDA) to derive insights from your data. Discover data visualization using ggplot2, unveiling the stories hidden within data. Lastly, get acquainted with RStudio, the robust Integrated Development Environment, enhancing your R programming journey. This workshop is your gateway to mastering R, catering to both novices and seasoned programmers.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="secfocus">Self Paced</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;"> Fine Tuning Existing LLMs and Embedding Models</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/odsc-instructor'> <img src='https://odsc.com/wp-content/uploads/2024/01/blue.png' alt='ODSC Instructor'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>ODSC Instructor</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;"></span></div></div><div class="etn-acccordion-contents "><p> This workshop explores the importance of fine-tuning Language and Embedding Models (LLMs). It highlights how embedding models are used to map natural language to vectors, crucial for pipelines with multiple models to adapt to specific data nuances. An example demonstrates fine-tuning an embedding model for legal text. The notebook discusses existing solutions and the hardware considerations, emphasizing GPU usage for large data. The practical part of the notebook shows the fine-tuning process of the """"distilroberta-base"""" model from the SentenceTransformer library. It utilizes the QQP_triplets dataset from Quora for training, designed around semantic meaning. The notebook prepares the data, sets up a DataLoader, and employs Triplet Loss to encourage the model to map similar data points closely while distancing dissimilar ones. It concludes by mentioning the training duration and resources needed for further improvements.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="firstfocus">Generative AI</span> <span class="secfocus">Self Paced</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">GenAI & LLMS: Building a Q&A Bot with LLMs, Vector Search, and LangChain</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/odsc-instructor'> <img src='https://odsc.com/wp-content/uploads/2024/01/blue.png' alt='ODSC Instructor'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>ODSC Instructor</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;"></span></div></div><div class="etn-acccordion-contents "><p> The workshop notebook delves into building a Question and Answering Bot based on a fixed knowledge base, covering the integration of concepts discussed in earlier notebooks about LLMs (Large Language Models) and prompting. Initially, it introduces a high-level architecture focusing on vector search—a method to retrieve similar items based on vector representations. The notebook explains the steps involved in vector search including vector representation, indexing, querying, similarity measurement, and retrieval, detailing various technologies used for vector search such as vector libraries, vector databases, and vector plugins. The example utilizes an Open Source vector database, Chroma, to index data and uses state of the union text data for the exercise. The notebook then transitions into the practical implementation, illustrating how text data is loaded, chunked into smaller pieces for effective vector search, and mapped into numeric vectors using the MPNetModel from the SentenceTransformer library via HuggingFace. Following this, the focus shifts to text generation where Langchain Chains are introduced. Chains, as described, allow for more complex applications by chaining several steps and models together into pipelines. A RetrievalQA chain is used to build a Q&A Bot application which utilizes an OpenAI chat model for text generation.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">AI Engineering</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">AI Copilots and Code Assistants</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/this-session-will-feature-a-leading-expert-in-the-field'> <img src='https://odsc.com/wp-content/uploads/2024/07/Screenshot-2024-07-22-at-19.03.02-Cropped.png' alt='This session will feature a leading expert in the field'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>This session will feature a leading expert in the field</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Check back for more information</span></div></div><div class="etn-acccordion-contents "><p> AI copilots and code assistants are revolutionizing the software development process by providing intelligent support to programmers. These AI-powered tools leverage advanced techniques like machine learning and natural language processing to enhance productivity, improve code quality, and accelerate development cycles.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Machine Learning</span> <span class="secfocus">Self Paced</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Introduction to Math for Data Science</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/odsc-instructor'> <img src='https://odsc.com/wp-content/uploads/2024/01/blue.png' alt='ODSC Instructor'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>ODSC Instructor</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;"></span></div></div><div class="etn-acccordion-contents "><p> Mathematics forms the backbone of data science, providing the essential tools for understanding, analyzing, and extracting insights from data. This session offers a foundational overview of key mathematical concepts indispensable for data scientists. Participants will explore probability, statistics, linear algebra, and calculus, gaining a solid grasp of their applications in real-world data challenges.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="firstfocus">Generative AI</span> <span class="secfocus">Self Paced</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">GenAI & LLMS: Introduction to Prompt Engineering</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/odsc-instructor'> <img src='https://odsc.com/wp-content/uploads/2024/01/blue.png' alt='ODSC Instructor'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>ODSC Instructor</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;"></span></div></div><div class="etn-acccordion-contents "><p> This workshop on Prompt Engineering explores the pivotal role of prompts in guiding Large Language Models (LLMs) like ChatGPT to generate desired responses. It emphasizes how prompts provide context, control output style and tone, aid in precise information retrieval, offer task-specific guidance, and ensure ethical AI usage. Through practical examples, participants learn how varying prompts can yield diverse responses, highlighting the importance of well-crafted prompts in achieving relevant and accurate text generation. Additionally, the workshop introduces temperature control to balance creativity and coherence in model outputs, and showcases LangChain, a Python library, to simplify prompt construction. Participants are equipped with practical tools and techniques to harness the potential of prompt engineering effectively, enhancing their interaction with LLMs across various contexts and tasks.</p></div></div></div></div></div></div> <!-- end repeatable item --> <!-- start repeatable item --><div class='etn-tab ' data-id='tab6740cdeeccc42-1'><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>9:40 am - 11:40 am</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Data Engineering</span> <span class="etn-schedule-location"> <span class="firstfocus">Machine Learning</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading active'><p style="width: 70%;float: left;">Introduction to Containers for Data Science / Data Engineering</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/michael-a-fudge'> <img src='https://odsc.com/wp-content/uploads/2024/03/Michael-A-Fudge.png' alt='Michael A Fudge'> </a></div></div> <i class="etn-icon etn-minus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Michael A Fudge</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Professor of Practice, MSIS Program Director at Syracuse University’s iSchool</span></div></div><div class="etn-acccordion-contents active"><p> In this hands on session, participants will learn how to leverage containers for data science / data engineering workflows. Containers allows us to bundle our application dependencies and configuration into an image, which can be more easily shared with others and deployed to the cloud. The session will explain how to use and build images, configure and run them and handle inter-dependencies between the product you're building and other services such as databases. Specifics covered: - how containers work - what are their advantages in data science / data engineering - finding images on repositories (docker hub / quay.io) - creating containers from the image and running it - exposing resources like ports and volumes - orchestration with docker-compose - building / configuring / customizing images to include your specific project dependencies - integrating your container with the visual studio code editors - containerizing dependent services like databases and integrating them with your project Source code from the workshop will be available for attendees on github.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:00 am - 1:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Deep Learning</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Enhancing AI Accuracy with Advanced Data Augmentation Techniques</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/duc-haba'> <img src='https://odsc.com/wp-content/uploads/2024/09/Duc-Haba.png' alt='Duc Haba'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Duc Haba</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Chief AI Officer at GenAI World Association</span></div></div><div class="etn-acccordion-contents "><p> In this presentation, I aim to delve into the transformative power of data augmentation in AI, particularly in deep learning and generative AI. I will explore the utility of over 150 functional, object-oriented methods and open-source libraries, leveraging real-world datasets to boost AI accuracy significantly. The session will include: - Practical demonstrations using Python Notebooks. - Showcasing the creation of customized charts and infographics. - Implementing effective data augmentation techniques. Key Features: Using geometric, photometric, and random erasing augmentation methods in image classification and segmentation. Explore advanced text augmentation techniques using AI models like BERT and GPT-2. Audio and tabular data augmentation demonstrations featuring real-world applications and innovative visualization methods. Analysis of biases and safe augmentation parameters enhances AI models' robustness and reliability.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:00 am - 1:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMOps & MLOps</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Accelerated Data Science DLI X-Lab</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/eric-phan'> <img src='https://odsc.com/wp-content/uploads/2024/10/Eric-Phan.png' alt='Eric Phan'> </a></div><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/kevin-lee'> <img src='https://odsc.com/wp-content/uploads/2024/10/Kevin-Lee_.png' alt='Kevin Lee'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Eric Phan</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">System Software Engineer at NVIDIA </span></div><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Kevin Lee</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Sr. Content Developer, Deep Learning at NVIDIA </span></div></div><div class="etn-acccordion-contents "><p> With datasets growing rapidly in volume, velocity, and veracity, the demand for more efficient data processing has never been higher. This workshop walks through how to apply open-source GPU accelerators from the NVIDIA RAPIDS project for common python data science workflows. In this training, you will: Speed up data manipulation tasks in pandas using cuDF, NVIDIA's GPU-accelerated DataFrame library, and explore advanced operations like grouping, sorting, and merging data. Discover how Polars enhances larger scale data processing with memory optimizations, lazy execution, and cuDF-powered GPU parallelization Explore GPU-accelerated feature engineering and ML modeling with XGBoost Perform graph analytics faster and more efficiently using nx-cugraph, a NetworkX backend that accelerates most of its popular algorithm</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>12:00 pm - 2:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Bootcamp</span> <span class="secfocus">Beginner</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">A Gentle Introduction to Vector Databases and Their Implementation</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/balaji-dhamodharan'> <img src='https://odsc.com/wp-content/uploads/2024/09/Balaji-Dhamodharan.png' alt='Balaji Dhamodharan'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Balaji Dhamodharan</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Data Science Leader at NXP Semiconductors</span></div></div><div class="etn-acccordion-contents "><p> In the rapidly advancing field of artificial intelligence, understanding the foundational technologies is key to leveraging their full potential. This session provides a gentle introduction to vector databases, an essential tool for enhancing the capabilities of large language models (LLMs). Designed for practitioners and enthusiasts alike, this presentation will cover the basics of vector databases, their applications, and an end-to-end hands-on implementation. We will start by demystifying what vector databases are, with clear definitions and simple explanations. Real-world examples of popular vector databases will be presented to illustrate their utility and versatility. The session will explain why vector databases are becoming indispensable in the AI landscape, particularly in improving the performance and scalability of LLMs. Attendees will gain a practical understanding of how vector databases work, including the processes involved in storing, retrieving, and managing data in high-dimensional vector spaces. The session will culminate in a comprehensive hands-on workshop, where participants will implement a vector database from scratch, gaining valuable experience in real-world application. Key topics to be covered include: - What is a Vector Database?: Introducing the concept and significance of vector databases in simple terms. - Examples of Vector Databases: Exploring popular vector databases such as FAISS, Milvus, and Pinecone, and their unique features. - Why need a Vector Database?: Understanding the benefits and applications of vector databases in enhancing the performance of LLMs and other AI systems. - How they work: Delving into the mechanics of vector databases, including indexing, similarity search, and data retrieval. - Hands-On Implementation: An end-to-end practical session where participants will set up and use a vector database, reinforcing the concepts learned. This session is designed to provide a solid foundation in vector databases, equipping attendees with both theoretical knowledge and practical skills. Whether you are a beginner or looking to deepen your understanding, this presentation will offer valuable insights and hands-on experience in the exciting world of vector databases. Join us to discover how vector databases can transform your approach to managing and leveraging data in AI applications.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:00 pm - 4:30 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Machine Learning</span> <span class="etn-schedule-location"> <span class="firstfocus">Data Visualization& Analysis</span> <span class="secfocus">Intermediate - Advanced</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Uncertainty Quantification: Approaches and Methods</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/brian-lucena-phd'> <img src='https://odsc.com/wp-content/uploads/2019/12/Brian-Lucena_.png' alt='Brian Lucena, PhD'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Brian Lucena, PhD</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Principal at Numeristical</span></div></div><div class="etn-acccordion-contents "><p> As machine learning models have become more present in our lives, there has been increasing attention on the reliability of these models. A major component of this is understanding the how uncertain the model is about its prediction. No model is exactly right 100% of the time, so we need methods and approaches by which we can quantify the level of uncertainty around a prediction. Approaches to uncertainty quantification (UQ) vary, and depend on the type of problem. For classification problems, the primary approach is probability calibration: making sure that the model outputs corresponding to each class "behaves well" as a probability. For regression problems, there are several different approaches. One can configure models to output an interval, rather than a single point prediction, along with a "coverage" value that specifies the probability that the interval covers the true value. The framework of Conformal Prediction provides theoretical guarantees around such interval predictions. Or one can use methods that output an entire conditional density for y given X. This is called probabilistic regression, or conditional density estimation. Several parametric and non-parametric approaches exist for this problem. This workshop will provide the theoretical context for these methods and then dive into real-world examples of their applications using Jupyter notebooks.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:00 pm - 5:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">AI Engineering</span> <span class="firstfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Mastering Web Data Acquisition Techniques</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/yohan-raju'> <img src='https://odsc.com/wp-content/uploads/2024/10/Yohan-Raju.png' alt='Yohan Raju'> </a></div><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/rafael-levi'> <img src='https://odsc.com/wp-content/uploads/2024/10/Rafael-Levi.png' alt='Rafael Levi'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Yohan Raju</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Solutions Consultant Expert at Bright Data</span></div><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Rafael Levi</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Senior Solutions Architecture Expert at Bright Data</span></div></div><div class="etn-acccordion-contents "><p> Join this workshop to explore the latest in web data collection, from serveless scraping to advanced APIs and AI-powered solutions. Learn about alternative web data acquisition approaches, get hands-on experience with Bright Data’s tool, and discover how to access structured data efficiently - on the fly and historical. We’ll cover the challenges of data access, ethical data collection, and real-world case studies, empowering you with practical tools and techniques for the evolving web data landscape.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>3:30 pm - 5:30 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="etn-schedule-location"> <span class="firstfocus">GenAI</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Building with Llama 3.2</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/varun-vontimitta'> <img src='https://odsc.com/wp-content/uploads/2024/10/Varun-Vontimitta.png' alt='Varun Vontimitta'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Varun Vontimitta</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">AI Partner Engineering Manager | Meta</span></div></div><div class="etn-acccordion-contents "><p> Llama 3.2, the latest version of the powerful AI model that has taken the developer community by storm. In this workshop, we will dive deep into the latest models and show you how to download, run inferences, call tools, fine-tune, and evaluate them. But that's not all! We will also cover the crucial topic of building and customizing safety into AI applications. As AI becomes more prevalent in our lives, it's essential to ensure that these systems are safe and ethical. Our expert speakers will share their insights and best practices on how to achieve this. Whether you're a seasoned developer or just starting out with AI, this workshop is perfect for anyone looking to take their skills to the next level. So, don't miss out on this opportunity to learn about the latest advancements in AI and how to build safe and responsible AI applications.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Advance LLMs - Agents, Parameter Efficient Fine-Tuning, and RAG</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/odsc-instructor'> <img src='https://odsc.com/wp-content/uploads/2024/01/blue.png' alt='ODSC Instructor'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>ODSC Instructor</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;"></span></div></div><div class="etn-acccordion-contents "><p> The workshop explores advanced topics in Large Language Models (LLMs) across three detailed sections. The first segment delves into LangChain Agents, showcasing the integration of LLMs with external systems to execute complex actions. The second part, """"Parameterized Fine-Tuning,"""" explores fine-tuning LLMs for specific tasks. In the final section, participants dive into """"Retrieval-Augmented Generation (RAG),"""" understanding how it merges retrieval and generation models to enhance language processing tasks. The section also elaborates on indexing content in a Retrieval Question-Answering chain for efficient information retrieval and showcases the use of Chroma, a vector database, for storing and managing high-dimensional vector data. Practical exercises, such as working with the CNN/DailyMail dataset, configuring chains, and prompt engineering, are peppered throughout, providing a hands-on learning experience on utilizing OpenAI and LangChain frameworks to tackle real-world problems. This workshop is a rich blend of theoretical knowledge and practical skills, aimed at harnessing the power of LLMs for varied applications.</p></div></div></div></div></div></div> <!-- end repeatable item --> <!-- start repeatable item --><div class='etn-tab ' data-id='tab6740cdeeccc42-2'><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>11:00 am - 1:00 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="etn-schedule-location"> <span class="firstfocus">GenAI</span> <span class="secfocus">Intermediate-Advanced</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading active'><p style="width: 70%;float: left;">LLMs from prototype to production - LLMOps, Prompt Engineering, and Moving LLMs to the Cloud</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/sinan-ozdemir'> <img src='https://odsc.com/wp-content/uploads/2023/07/Sinan-Ozdemir_.png' alt='Sinan Ozdemir'> </a></div></div> <i class="etn-icon etn-minus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Sinan Ozdemir</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">AI & LLM Expert | Author | Founder + CTO at LoopGenius</span></div></div><div class="etn-acccordion-contents active"><p> This session is an extensive guide to moving LLMs to production from the prototype phase, with a focus on the critical aspects of LLMOps, prompt engineering, and cloud-based deployments. Learn to fine-tune and optimize LLMs such as GPT, Llama, and BERT for industry-specific applications while ensuring efficient and scalability. Key areas of focus include quantization for reducing model size without compromising too much performance, distillation techniques to streamline models for faster inference, and best practices for cloud-based deployment of models. Attendees will also gain hands-on experience in advanced prompt engineering, exploring techniques such as few-shot learning and chain-of-thought prompting to enhance LLM performance and consistency across various tasks and models. The session will also cover essential strategies for task evaluation, helping participants to identify areas for improvement through model monitoring and tight feedback integration. By the end of this session, attendees will have the practical skills to fine-tune LLMs, perform model quantization and distillation, and deploy optimized LLMs to the cloud, ensuring high efficiency and performance in real-world applications.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>12:25 pm - 2:25 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMs</span> <span class="secfocus">Intermediate - Advanced</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Building a Multimodal AI Assistant: Build an AI Application Using Advanced RAG for Cross-Modal Data Retrieval</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/suman-debnath'> <img src='https://odsc.com/wp-content/uploads/2024/10/Suman-Debnath.png' alt='Suman Debnath'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Suman Debnath</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Principal AI/ML Advocate at Amazon Web Services</span></div></div><div class="etn-acccordion-contents "><p> In this session, we explore the technical architecture behind creating a multimodal AI assistant using advanced Retrieval-Augmented Generation (RAG) techniques integrated with LlamaIndex for efficient data retrieval across diverse sources. We will discuss how to address the limitations of native RAG models, including challenges with incomplete data, reasoning mismatches, and handling multimodal inputs like text, tables, and images. By leveraging LlamaIndex, along with visual language models and embedding techniques, we enable cross-modal understanding and more accurate information retrieval. Attendees will learn how to utilize LlamaIndex for structuring and indexing large datasets, combine it with LangChain-based embeddings, and implement query decomposition and fusion strategies to enhance AI performance. This session is ideal for developers and data scientists looking to build robust AI systems capable of reasoning and retrieval across varied data types and formats.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:00 pm - 4:30 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">Deep Learning</span> <span class="etn-schedule-location"> <span class="firstfocus">AI Engineering</span> <span class="secfocus">All Levels</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Going From Unstructured Data to Vector Similarity Search</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/steven-pousty-phd'> <img src='https://odsc.com/wp-content/uploads/2020/02/Steven-Pousty.png' alt='Steven Pousty, PhD'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Steven Pousty, PhD</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Principal and Founder at Tech Raven Consulting</span></div></div><div class="etn-acccordion-contents "><p> One of the key concepts used in AI modeling is the storage and query of vectors. This workshop will start with 2 examples of unstructured data, images and journal abstracts. Participants will then work this data all the way through to a usable data store with an application on top it. We will cover things such as transformers, embeddings, HuggingFace, choosing a vectorization model, and effective query composition. The python code needed to do this is quite easy to understand. If you want to work on your own laptop, there will be prerequisites published beforehand. If not, we will use a cloud-hosted environment for you to do all your work. After this workshop you should be able to understand some of the concepts being used in these new AI architectures and have a better grasp on the work involved with building “AI” applications. Software needs/requirements: • A web browser • A github account • The ability to read some simple python and SQL code Key learnings/takeaways • How to turn unstructured data into knowledge • What are vectors (also called embeddings) • Foundational models versus creating your own • How you can make your own enhacements to foundational models • How to choose an appropriate vector size for your use case • How to lead vector data into a vector data store • What is the HNSW index and why you should care • Queries that are appropriate for your data Benefits of attending the workshop: • Practical hands on work with some of the latest AI technology • A better grasp of the process of using it with your own data • A better idea of how to integrate these technologies into your ◦ Data flows ◦ App Architectures ◦ App development</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-time'>2:35 pm - 4:35 pm</span> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">LLMOPs</span> <span class="etn-schedule-location"> <span class="firstfocus">Data Engineering</span> <span class="secfocus">Intermediate-Advanced</span> </span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">LLMOps Infrastructure for Production-grade RAG Applications</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/niels-bantilan'> <img src='https://odsc.com/wp-content/uploads/2024/09/Niels-Bantilan.png' alt='Niels Bantilan'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Niels Bantilan</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Chief ML Engineer at Union.ai</span></div></div><div class="etn-acccordion-contents "><p> So you’ve built and deployed a RAG proof-of-concept at your organization with an off-the-shelf framework in a few days, and it looks like it’s working… now what? Just as in previous generations of AI and ML, the journey of a RAG application doesn’t end once you’ve deployed a model: it requires constant iteration to maintain and improve its performance. Unlike traditional ML deployments, however, AI applications like RAG involve more complex operational and infrastructure requirements such as maintaining pipeline logic, vector stores, evaluation datasets, and LLM model hosting. In this workshop, you’ll learn what it takes to maintain and iterate on RAG applications from prototype to production by building a chat assistant using the Union orchestrator. The workshop is structured in three parts: - Creating a baseline RAG application: In this section, we’ll look at the anatomy of a basic RAG pipeline and understand what building blocks you need to build robust and reliable AI applications. This includes container-native DAGs, built-in type-safety and data validation, versioning and management of vector stores and models, and reusable container environments. - Using LLM-as-a-judge to bootstrap an evaluation dataset: In this section, we’ll build an evaluation dataset using an LLM so that we can begin to iterate on and improve on the baseline pipeline. Here we’ll learn how to use an LLM to create synthetic questions based on ground truth documents and assess the resulting Q&A pairs using a critic LLM. We’ll also introduce a human-in-the-loop element to manually check samples of the synthetic evaluation dataset so that we can create few-shot examples of Q&A judgements to help the critic assess synthetic data. - Hyperparameter optimization: Using the evaluation dataset, we will perform hyperparameter optimization, which in this case are parameters like the generation prompt, text chunking procedure, and top-k value for retrieving documents. Using well-established train/test validation techniques from traditional hyperparameter optimization, we can escape “vibe-based” pipeline evaluation and instead improve RAG performance more objectively. Attendees will take away the core concepts and techniques required to systematically improve their RAG applications while adopting best software engineering practices.</p></div></div></div></div></div></div> <!-- end repeatable item --></div></div> <!-- schedule tab end --><br /></div><div id="fragment-4-499863582" class="clearfix be-tab-content"><p><b> <!-- schedule tab start --><div class="schedule-tab-wrapper etn-tab-wrapper schedule-tab-2"><ul class='etn-nav'><li> <a href='#' class='etn-tab-a etn-active' data-id='tab6740cdeeda80a-0'> <span class='etn-date'>30 Oct</span> <span class=etn-day>Day 1 - Day 3</span> </a></li></ul><div class='etn-tab-content clearfix etn-schedule-wrap'> <!-- start repeatable item --><div class='etn-tab tab-active' data-id='tab6740cdeeda80a-0'><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">VIP Passes</span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading active'><p style="width: 70%;float: left;">VIP & Speaker Breakfast</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"></div> <i class="etn-icon etn-minus"></i></h4><div class='etn-schedule-speaker-item'></div><div class="etn-acccordion-contents active"><p> Kickstart your day with a delicious breakfast and stimulating conversation! Join us for an informal gathering of data science professionals and enthusiasts. This is your chance to connect with Ai Experts, share ideas, and explore new opportunities in a relaxed setting.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Passes</span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Roullette Talks</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'></div><div class="etn-acccordion-contents "><p> Get ready to be surprised, entertained, and informed! Join AI experts as they test their knowledge and their presentation skills with randomly assigned slides.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">VIP Passes</span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">VIP & Speakers Round Table Lunch</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'></div><div class="etn-acccordion-contents "><p> A rare opportunity to join a roundtable discussion with fellow VIPs and renowned speakers. Share your expertise, spark innovation, and leave inspired by the collective wisdom of the ODSC community.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Passes</span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">AI Solution Showcase</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'></div><div class="etn-acccordion-contents "><p> Step into the future of AI at ODSC’s AI Solution Showcase Expo Hall, a vibrant hub of innovation and opportunity. Our Expo Hall is designed to ignite your curiosity and foster connections, featuring an impressive lineup of AI solution showcase partners ready to demonstrate their cutting-edge technologies and services. Relax and network in our specially designed hangout areas, perfect for conversations and sharing insights. Whether you’re seeking practical AI solutions for your business, or simply eager to see what the future holds, the ODSC Expo Hall is your gateway to the exciting world of AI. Get ready to be inspired, challenged, and engaged.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Passes</span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Speaker's Office Hours</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'></div><div class="etn-acccordion-contents "><p> ODSC speakers and instructors are some of the best and brightest in the AI industry. At the exclusive office hours events, you’ll have a once-in-a-lifetime chance to meet AI pioneers, ask them questions, and learn about their experiences in a casual setting.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Passes</span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">VIP & Speakers Networking Reception</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'></div><div class="etn-acccordion-contents "><p> Don’t miss out on the vibrant after-session scene at ODSC! Our four evening receptions offer something for everyone. Mingle with fellow attendees and industry leaders at the Welcome Reception (Tuesday). The Main Networking Reception (Wednesday) is a central hub to unwind, share insights, and forge new connections. Wrap up the conference on a high note at the Wrap-up Reception (Thursday) and solidify the connections that will fuel your AI journey.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Passes</span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Book Signing & Author Meet-and-Greet</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'></div><div class="etn-acccordion-contents "><p> Connect with leading voices in data science and technology at our exclusive book signing event. Meet renowned authors, get your books personalized, and engage in insightful conversations about their latest works.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Passes</span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Daily Meetups</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'></div><div class="etn-acccordion-contents "><p> At ODSC, every break is an opportunity to engage. Between sessions, we invite attendees to connect over lunch and coffee in a relaxed, informal setting. These gatherings are designed not just to nourish the body, but also to foster rich exchanges of stories and insights, allowing participants to share experiences, discuss the latest trends in AI and data science, and explore collaborative opportunities.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Passes</span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Ai Startup Solution Expo Hall</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'></div><div class="etn-acccordion-contents "><p> Step into the future of innovation at our Startup Expo Hall. Discover a vibrant showcase of cutting-edge technologies, groundbreaking ideas, and the next generation of industry disruptors. Meet the brilliant minds behind these startups, explore their products and services, and witness the potential to transform our world.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Passes</span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Startup Pitch & Networking</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'></div><div class="etn-acccordion-contents "><p> Witness the future of data science unfold at our Startup Pitches & Networking Reception. Join us for an exciting evening where innovative startups take the stage to pitch their groundbreaking ideas and technologies.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Passes</span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Data After Dark</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'></div><div class="etn-acccordion-contents "><p> Unwind after a stimulating day at ODSC with fellow AI and data science enthusiasts! Join us for a casual drinks, happy hours, and AI Travia. It’s a great opportunity to network in a relaxed setting, share insights, and forge new connections. Our venues include AI Trivia Night, Live Music & Networking, and Happy Hour Specials. Location and times available on Day-of Schedule.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Passes</span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Virtual Hackathon: Build Highly Accurate, Custom Generative AI Models with NVIDIA NeMo</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'></div><div class="etn-acccordion-contents "><p> Dive into the full AI workflow over an exhilarating weekend challenge! Our virtual hackathon, led by the NVIDIA NeMo team, will guide you through the end-to-end journey of customizing large language models(LLMs), from accelerated data processing to fine-tuning and evaluation. This hackathon offers invaluable hands-on experience in building highly accurate customized models, equipping you with the tools and knowledge to excel in real-world AI development. Join us for a weekend of learning, collaboration, and cutting-edge innovation!</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Passes</span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">In-Person Hackathon at ODSC West: Accelerate Machine Learning/AI Challenge</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'></div><div class="etn-acccordion-contents "><p> For those ready to collaborate and innovate in real-time, join us at ODSC West for an in-person hackathon experience! Explore GPU-powered machine learning and AI techniques to accelerate popular tools including Pandas or Polars in your solutions, and experience the speed and power of NVIDIA’s RAPIDS-accelerated Python libraries. This hackathon offers the opportunity to elevate your AI and ML workflows by optimizing performance, scaling solutions, and accelerating innovation.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Passes</span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">“Make the Jump” Dinner</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'></div><div class="etn-acccordion-contents "><p> At this intimate gathering, you’ll be part of a hand-picked group of like-minded professionals. There you will receive feedback on your ideas from early-stage VCs and fellow practitioners, connect with potential co-founders, gain valuable advice on launching your startup, and connect with peers. The dinner will take place on Thursday, October 10th, in San Francisco.</p></div></div></div></div></div><div class='etn-single-schedule-item etn-row'><div class='etn-schedule-info etn-col-lg-3 etn-col-sm-3'> <span class='etn-schedule-location'> <i class='etn-icon etn-location'></i> <span class="etn-schedule-location"> <span class="firstfocus">All Passes</span> </span></div><div class='etn-schedule-content etn-col-lg-9 etn-col-sm-9'><div class="etn-accordion-wrap etn-schedule-content-wrap xx"><div class="etn-content-item"><h4 class='etn-title etn-accordion-heading '><p style="width: 70%;float: left;">Women in Data Science Ignite: Sharing Insights and Networking Session</p><div class='etn-schedule-speaker-item' style="width: 30%;float: right;"><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/sweta-sinha'> <img src='https://odsc.com/wp-content/uploads/2024/10/Sweta-Sinha.png' alt='Sweta Sinha'> </a></div><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/micaela-kaplan'> <img src='https://odsc.com/wp-content/uploads/2024/10/Micaela-Kaplan.png' alt='Micaela Kaplan'> </a></div><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/kristy-hollingshead-phd'> <img src='https://odsc.com/wp-content/uploads/2024/10/Kristy-Hollingshead.png' alt='Kristy Hollingshead, PhD'> </a></div><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/maria-lupetini'> <img src='https://odsc.com/wp-content/uploads/2024/10/Maria-Lupetini.png' alt='Maria Lupetini'> </a></div><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/swagata-ashwani'> <img src='https://odsc.com/wp-content/uploads/2024/10/Swagata-Ashwani.png' alt='Swagata Ashwani'> </a></div><div class='etn-schedule-single-speaker' style="float:right"> <a href='https://odsc.com/blog/speaker/dr-shelby-heinecke'> <img src='https://odsc.com/wp-content/uploads/2024/10/Shelby-Heinecke.png' alt='Dr. Shelby Heinecke'> </a></div></div> <i class="etn-icon etn-plus"></i></h4><div class='etn-schedule-speaker-item'><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Sweta Sinha</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Executive Director at JP Morgan Chase</span></div><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Micaela Kaplan</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Machine Learning Evangelist at HumanSignal</span></div><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Kristy Hollingshead, PhD</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Senior Data Science Lead at Further</span></div><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Maria Lupetini</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">CEO and Chief Data Scientist at InfoMaker Inc</span></div><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Swagata Ashwani</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Principal Data Scientist/Data Science Lead at Boomi</span></div><div class='etn-schedule-single-speaker'> <span class='etn-schedule-speaker-title'>Dr. Shelby Heinecke</span> <span class='etn-schedule-speaker-designation' style=" margin-bottom: 15px;">Senior AI Research Manager at Salesforce</span></div></div><div class="etn-acccordion-contents "><p></p></div></div></div></div></div></div> <!-- end repeatable item --></div></div> <!-- schedule tab end --></b></p></div></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div><div class="tatsu-overlay tatsu-column-overlay tatsu-animate-none" ></div></div><style>.tatsu-row > .tatsu-SJ8cp25Phh.tatsu-column{width: 100%;}.tatsu-SJ8cp25Phh.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: normal;}.tatsu-SJ8cp25Phh > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-SJ8cp25Phh > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-SJ8cp25Phh > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-SJ8cp25Phh > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width:1377px) {.tatsu-row > .tatsu-SJ8cp25Phh.tatsu-column{width: 100%;}}@media only screen and (min-width:768px) and (max-width: 1024px) {.tatsu-row > .tatsu-SJ8cp25Phh.tatsu-column{width: 100%;}}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-SJ8cp25Phh.tatsu-column{width: 100%;}}</style></div></div></div></div><div class="tatsu-section-background-wrap"><div class = "tatsu-section-background" ></div></div><div class="tatsu-overlay tatsu-section-overlay"></div><style>.tatsu-HJqanqwh3.tatsu-section{background-color: rgba(0,0,0,1);}.tatsu-HJqanqwh3 .tatsu-section-pad{padding: 0px 0px 30px 0px;}.tatsu-HJqanqwh3 .tatsu-section-offset-wrap{transform: translateY(-0px);}.tatsu-HJqanqwh3 > .tatsu-bottom-divider{z-index: 9999;}.tatsu-HJqanqwh3 > .tatsu-top-divider{z-index: 9999;}.tatsu-HJqanqwh3 .tatsu-section-overlay{mix-blend-mode: normal;}</style></div><div class="tatsu-hryhe4mngk7kry0y tatsu-section tatsu-bg-overlay tatsu-hide-tablet tatsu-hide-mobile tatsu-clearfix" data-title="D - Colocated" data-headerscheme="background--dark"><div class='tatsu-section-pad clearfix' data-padding='{"d":"0px 0px 55px 0px"}' data-padding-top='0px'><div class="tatsu-row-wrap tatsu-wrap tatsu-row-one-col tatsu-row-has-one-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-hryhe4mnsx2gc2ah" ><div class="tatsu-row " ><div class="tatsu-column tatsu-bg-overlay tatsu-one-col tatsu-column-image-none tatsu-column-effect-none tatsu-hryhe4mnwh5rty73" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class="tatsu-module tatsu-text-block-wrap tatsu-hryhe4mnyy45ricl "><div class="tatsu-text-inner tatsu-align-center clearfix" ><style>.tatsu-hryhe4mnyy45ricl.tatsu-text-block-wrap .tatsu-text-inner{width: 100%;text-align: left;color: rgba(255,255,255,1) ;}.tatsu-hryhe4mnyy45ricl .tatsu-text-inner *{color: rgba(255,255,255,1) ;}</style><h2 style="text-align: center;"><span style="color: #ffffff;"><strong><b>Co-located @ ODSC West</b></strong></span></h2><h4 style="text-align: center;"><span style="color: #ffffff;">Discover How to do More with data at the GEN Ai X and Data Engineering Summits. Access is included with an ODSC Silver Pass or above</span></h4></div></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div><div class="tatsu-overlay tatsu-column-overlay tatsu-animate-none" ></div></div><style>.tatsu-row > .tatsu-hryhe4mnwh5rty73.tatsu-column{width: 100%;}.tatsu-hryhe4mnwh5rty73.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: normal;}.tatsu-hryhe4mnwh5rty73 > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-hryhe4mnwh5rty73 > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-hryhe4mnwh5rty73 > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-hryhe4mnwh5rty73 > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width:1377px) {.tatsu-row > .tatsu-hryhe4mnwh5rty73.tatsu-column{width: 100%;}}@media only screen and (min-width:768px) and (max-width: 1024px) {.tatsu-row > .tatsu-hryhe4mnwh5rty73.tatsu-column{width: 100%;}}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-hryhe4mnwh5rty73.tatsu-column{width: 100%;}}</style></div></div></div><div class="tatsu-row-wrap tatsu-row-full-width tatsu-row-has-one-half tatsu-row-has-two-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-hryhe4mo0c2qn90l" ><div class="tatsu-row " ><div class="tatsu-column tatsu-bg-overlay tatsu-one-half tatsu-column-align-top tatsu-column-image-none tatsu-column-effect-none tatsu-hryhe4mo3v3yjvxp" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class="tatsu-single-image tatsu-module align-right tatsu-image-lazyload tatsu-external-image tatsu-hryhe4mortatu3ok " ><div class="tatsu-single-image-inner " style="" ><div class = "tatsu-single-image-padding-wrap" style = "" ></div><a href = "/california/aix-west/" ><img class = "tatsu-gradient-border" data-src = "https://odsc.com/wp-content/uploads/2024/06/Co-Located-Events-Banners-GEN-AI-X-WEST-2024-1.png" alt =" " src ="data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAQAAAC1HAwCAAAAC0lEQVR42mNkYAAAAAYAAjCB0C8AAAAASUVORK5CYII=" /></a></div><style>.tatsu-hryhe4mortatu3ok .tatsu-single-image-inner{border-style: solid;max-width: 79%;}.tatsu-hryhe4mortatu3ok.tatsu-single-image{transform: translate3d(0px,0px, 0);}</style></div><div class="tatsu-module tatsu-normal-button tatsu-button-wrap align-block block-center tatsu-hryhe4mot84dqctk button-transform "><a class="tatsu-shortcode x-largebtn tatsu-button left-icon rounded tatsu-animate bg-animation-none " href="/california/aix-west/" style= "" data-animation="wiggle" data-animation-delay="743" data-animation-duration="1552" aria-label="LEARN MORE " data-gdpr-atts={} target="_blank">LEARN MORE </a><style>.tatsu-hryhe4mot84dqctk .tatsu-button{background-color: rgba(255,98,0,1);color: rgba(255,255,255,1) ;border-color: rgba(187,104,13,1); }.tatsu-hryhe4mot84dqctk.tatsu-normal-button{margin: 0% 0px 0px 0%;}</style></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div><div class="tatsu-overlay tatsu-column-overlay tatsu-animate-none" ></div></div><style>.tatsu-row > .tatsu-hryhe4mo3v3yjvxp.tatsu-column{width: 50%;}.tatsu-hryhe4mo3v3yjvxp.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: normal;}.tatsu-hryhe4mo3v3yjvxp > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-hryhe4mo3v3yjvxp > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-hryhe4mo3v3yjvxp > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-hryhe4mo3v3yjvxp > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width:1377px) {.tatsu-row > .tatsu-hryhe4mo3v3yjvxp.tatsu-column{width: 50%;}}@media only screen and (min-width:768px) and (max-width: 1024px) {.tatsu-row > .tatsu-hryhe4mo3v3yjvxp.tatsu-column{width: 50%;}}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-hryhe4mo3v3yjvxp.tatsu-column{width: 100%;}}</style></div><div class="tatsu-column tatsu-column-no-bg tatsu-one-half tatsu-column-image-none tatsu-column-effect-none tatsu-hryhe4mov2fade3x" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class="tatsu-single-image tatsu-module align-left tatsu-external-image tatsu-hryhe4mpcdeohoic " ><div class="tatsu-single-image-inner " style="" ><div class = "tatsu-single-image-padding-wrap" style = "" ></div><a href = "https://summit.ai/" target = "_blank" ><img class = "tatsu-gradient-border" src = "https://odsc.com/wp-content/uploads/2024/06/Co-Located-Events-Banners-DE-WEST-2024-1.png" alt =" " /></a></div><style>.tatsu-hryhe4mpcdeohoic .tatsu-single-image-inner{border-style: solid;max-width: 79%;}.tatsu-hryhe4mpcdeohoic.tatsu-single-image{transform: translate3d(0px,0px, 0);}</style></div><div class="tatsu-module tatsu-normal-button tatsu-button-wrap align-block block-center tatsu-hryhe4mpdlc24d79 button-transform "><a class="tatsu-shortcode x-largebtn tatsu-button left-icon rounded tatsu-animate bg-animation-none " href="https://summit.ai/" style= "" data-animation="wiggle" data-animation-delay="2145" data-animation-duration="1632" aria-label="LEARN MORE " data-gdpr-atts={} target="_blank">LEARN MORE </a><style>.tatsu-hryhe4mpdlc24d79 .tatsu-button{background-color: rgba(65,117,5,1);color: rgba(255,255,255,1) ;border-color: rgba(187,104,13,1); }.tatsu-hryhe4mpdlc24d79.tatsu-normal-button{margin: 0% 0px 0px -20%;}</style></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div></div><style>.tatsu-row > .tatsu-hryhe4mov2fade3x.tatsu-column{width: 50%;}.tatsu-hryhe4mov2fade3x.tatsu-column{margin: 0px 0px 0px 100px !important;}.tatsu-hryhe4mov2fade3x.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: none;}.tatsu-hryhe4mov2fade3x > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-hryhe4mov2fade3x > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-hryhe4mov2fade3x > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-hryhe4mov2fade3x > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width:1377px) {.tatsu-row > .tatsu-hryhe4mov2fade3x.tatsu-column{width: 50%;}}@media only screen and (min-width:768px) and (max-width: 1024px) {.tatsu-row > .tatsu-hryhe4mov2fade3x.tatsu-column{width: 50%;}}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-hryhe4mov2fade3x.tatsu-column{width: 100%;}}</style></div></div></div></div><div class="tatsu-section-background-wrap"><div class = "tatsu-section-background" ></div></div><div class="tatsu-overlay tatsu-section-overlay"></div><style>.tatsu-hryhe4mngk7kry0y.tatsu-section{background-color: rgba(0,0,0,1);}.tatsu-hryhe4mngk7kry0y .tatsu-section-pad{padding: 0px 0px 55px 0px;}.tatsu-hryhe4mngk7kry0y .tatsu-section-offset-wrap{transform: translateY(-0px);}.tatsu-hryhe4mngk7kry0y > .tatsu-bottom-divider{z-index: 9999;}.tatsu-hryhe4mngk7kry0y > .tatsu-top-divider{z-index: 9999;}.tatsu-hryhe4mngk7kry0y .tatsu-section-overlay{mix-blend-mode: normal;}</style></div><div class="tatsu-gw02y9357ubnd3dh tatsu-section tatsu-bg-overlay tatsu-hide-tablet tatsu-hide-mobile tatsu-clearfix" data-title="" data-headerscheme="background--dark"><div class='tatsu-section-pad clearfix' data-padding='{"d":"0px 0px 1px 0px"}' data-padding-top='0px'><div class="tatsu-row-wrap tatsu-wrap tatsu-row-one-col tatsu-row-has-one-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-HJgNY5meRq" ><div class="tatsu-row " ><div class="tatsu-column tatsu-bg-overlay tatsu-one-col tatsu-column-image-none tatsu-column-effect-none tatsu-SJNY57g09" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class = " tatsu-hiyn4b4d7r6ukt1k accordion-wrap oshine-module" ><style>.tatsu-hiyn4b4d7r6ukt1k{margin: 25px 0px 0px 0px;}</style><div class = "accordion" data-collapsed="1"><h3 class="accordion-head no-bg tatsu-hiyn4b4da294ga6b" style="">Where do I see Schedule In-Brief?</h3><div><h5>The ODSC Schedule overview is available on <a href="https://odsc.com/california/schedule-overview/">this page</a>.</h5></div><h3 class="accordion-head no-bg tatsu-rynz4P6_A" style="">WHICH SESSIONS ARE INCLUDED IN MY PASS?</h3><div><ul><li>ODSC Talks/Keynotes schedule includes Tuesday – Thursday. In-person sessions are available to<b> Silver, Gold, Platinum, Mini-Bootcamp, and VIP Pass</b> holders. Business talks are available to <b>Ai x Pass </b>holders. Virtual Sessions are available to <b>Virtual Premium, Virtual Platinum & Virtual Mini-Bootcamp</b> pass holders.</li><li class="li1">ODSC Trainings are scheduled from Monday – Wednesday. In-person sessions are available to <b>Platinum, Mini-Bootcamp, and VIP Pas</b>s holders. Virtual Sessions are available to V<b>irtual Platinum & Virtual Mini-Bootcamp pass holders.</b></li><li class="li1">ODSC Workshop/Tutorials are scheduled from Tuesday – Thursday. All in-person sessions are available to <b>VIP, Platinum, Mini-Bootcamp and Gold pass</b> holders. Silver Pass holders can attend only on Wednesday and Thursday. Virtual Sessions are available for V<b>irtual Premium, Virtual Platinum & Virtual Mini-Bootcamp pass holders.</b></li><li class="li1">ODSC Bootcamp Sessions are scheduled VIRTUALLY on Monday, as pre-conference training. They are ONLY available for <b>Mini-Bootcamp, and VIP Pass and Virtual Mini-Bootcam</b>p holders.</li></ul><p>Only virtual sessions are recorded. If you have a virtual pass, please note that we will not live-stream any in-person sessions. All Self-paced sessions are also available virtually </p></div><h3 class="accordion-head no-bg tatsu-hiyn4b4ddb3djgmp" style="">Other FAQS</h3><div><p>Please check the FAQ page <a href="https://odsc.com/california/faq/">HERE</a> for more information. It includes the in-person and virtual FAQs same as for Virtual Platform guidelines.</p></div><h3 class="accordion-head no-bg tatsu-hiyn4b4ddq21kjxl" style="">Contact support</h3><div><p>Our Support Team is always ready to assist you with any questions at <a href="mailto:info@odsc.com">info@odsc.com</a></p></div></div></div><div class="tatsu_testimonial_wrap tatsu-hiymtvb915e117mb clearfix bubble_center " ><div class="tatsu_testimonial_wrap"><div class="tatsu_testimonial_inner_wrap"><i class="tatsu-icon icon-quote"></i><div class="tatsu_testimonial_content"><div class="tatsu_testimonial_description">Please Note: In-person attendees will have access to virtual sessions. All sessions will fill up on a first-come-first-served basis.<br /> If you have a virtual pass, please remember that we will not live-stream any in-person sessions. Only virtual sessions will be recorded.</div></div></div></div><div class="tatsu_testimonial_info_wrap clearfix"><div class="tatsu_testimonial_info"><h6 class="tatsu_testimonial_author">All sessions are scheduled in the PDT time zone (Pacific Time)</h6></div></div><style>.tatsu-hiymtvb915e117mb .icon-quote, .tatsu-hiymtvb915e117mb .tatsu_testimonial_description{color: #ffffff ;}.tatsu-hiymtvb915e117mb .tatsu_testimonial_inner_wrap{border-color: rgba(74,74,74,1); }.tatsu-hiymtvb915e117mb .tatsu_testimonial_content{background-color: rgba(74,74,74,1);}.tatsu-hiymtvb915e117mb .tatsu_testimonial_author{color: rgba(0,14,20,1) ;}.tatsu-hiymtvb915e117mb .tatsu_testimonial_role{color: rgba(239,247,243,1) ;}.tatsu-hiymtvb915e117mb{margin: 15px 0px 0px 0px;}</style></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div><div class="tatsu-overlay tatsu-column-overlay tatsu-animate-none" ></div></div><style>.tatsu-row > .tatsu-SJNY57g09.tatsu-column{width: 100%;}.tatsu-SJNY57g09.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: normal;}.tatsu-SJNY57g09 > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-SJNY57g09 > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-SJNY57g09 > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-SJNY57g09 > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width:1377px) {.tatsu-row > .tatsu-SJNY57g09.tatsu-column{width: 100%;}}@media only screen and (min-width:768px) and (max-width: 1024px) {.tatsu-row > .tatsu-SJNY57g09.tatsu-column{width: 100%;}}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-SJNY57g09.tatsu-column{width: 100%;}}</style></div></div></div></div><div class="tatsu-section-background-wrap"><div class = "tatsu-section-background" ></div></div><div class="tatsu-overlay tatsu-section-overlay"></div><style>.tatsu-gw02y9357ubnd3dh .tatsu-section-pad{padding: 0px 0px 1px 0px;}.tatsu-gw02y9357ubnd3dh .tatsu-section-offset-wrap{transform: translateY(-0px);}.tatsu-gw02y9357ubnd3dh > .tatsu-bottom-divider{z-index: 9999;}.tatsu-gw02y9357ubnd3dh > .tatsu-top-divider{z-index: 9999;}.tatsu-gw02y9357ubnd3dh .tatsu-section-overlay{mix-blend-mode: normal;}</style></div><div class="tatsu-Bkn-6cP23 tatsu-section tatsu-bg-overlay tatsu-hide-mobile tatsu-hide-tablet tatsu-hide-laptop tatsu-hide-desktop tatsu-clearfix" data-title="First 50 - Talks" data-headerscheme="background--dark"><div class='tatsu-section-pad clearfix' data-padding='{"d":"90px 0px 90px 0px"}' data-padding-top='90px'><div class="tatsu-row-wrap tatsu-wrap tatsu-row-one-col tatsu-row-has-one-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-Syl2Zp9Dh2" ><div class="tatsu-row " ><div class="tatsu-column tatsu-bg-overlay tatsu-one-col tatsu-column-image-none tatsu-column-effect-none tatsu-HkWhZpqw2n" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class="tatsu-module tatsu-text-block-wrap tatsu-HyznbT5vhh "><div class="tatsu-text-inner tatsu-align-center clearfix" ><style>.tatsu-HyznbT5vhh.tatsu-text-block-wrap .tatsu-text-inner{width: 100%;text-align: left;background-color: rgba(0,0,0,1);}</style><p style="text-align: center;"><span style="color: #ffffff; font-size: 24pt;">Past Talks & Panels</span></p></div></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div><div class="tatsu-overlay tatsu-column-overlay tatsu-animate-none" ></div></div><style>.tatsu-row > .tatsu-HkWhZpqw2n.tatsu-column{width: 100%;}.tatsu-HkWhZpqw2n.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: normal;}.tatsu-HkWhZpqw2n > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-HkWhZpqw2n > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-HkWhZpqw2n > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-HkWhZpqw2n > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width:1377px) {.tatsu-row > .tatsu-HkWhZpqw2n.tatsu-column{width: 100%;}}@media only screen and (min-width:768px) and (max-width: 1024px) {.tatsu-row > .tatsu-HkWhZpqw2n.tatsu-column{width: 100%;}}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-HkWhZpqw2n.tatsu-column{width: 100%;}}</style></div></div></div><div class="tatsu-row-wrap tatsu-wrap tatsu-row-has-one-half tatsu-row-has-two-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-ByQhWT5w22" ><div class="tatsu-row " ><div class="tatsu-column tatsu-bg-overlay tatsu-one-half tatsu-column-image-none tatsu-column-effect-none tatsu-Bk4hZp5P23" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class = "tatsu-module tatsu-icon_card tatsu-ryS3b6cP3h tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-ryS3b6cP3h .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-ryS3b6cP3h .tatsu-icon_card-title, .tatsu-ryS3b6cP3h .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-ryS3b6cP3h .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-ryS3b6cP3h.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Chelsea Finn - Assistant Professor | Stanford University </a></div><div class = "tatsu-icon_card-caption body"><p>Keynote: <span data-sheets-value="{" data-sheets-userformat="{">Neural Networks Make Stuff up. What Should We do About it?</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-r19H-gCA2 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-r19H-gCA2 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-r19H-gCA2 .tatsu-icon_card-title, .tatsu-r19H-gCA2 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-r19H-gCA2 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-r19H-gCA2.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Yaron Haviv - Co-Founder and CTO | Iguazio (Acquired by Mckinsey) </a></div><div class = "tatsu-icon_card-caption body"><p>Track Keynote: <span data-sheets-value="{" data-sheets-userformat="{">Implementing Gen AI in Practice</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-SyI2-6cvh3 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-SyI2-6cvh3 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-SyI2-6cvh3 .tatsu-icon_card-title, .tatsu-SyI2-6cvh3 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-SyI2-6cvh3 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-SyI2-6cvh3.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Lukas Biewald - CEO and Co-founder | Weights & Biases </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Talk: Understanding the Landscape of Large Models</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-r1wnZaqPhh tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-r1wnZaqPhh .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-r1wnZaqPhh .tatsu-icon_card-title, .tatsu-r1wnZaqPhh .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-r1wnZaqPhh .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-r1wnZaqPhh.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Michael Auli - Principal Research Scientist | Director | FAIR | Meta AI </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Talk: General and Efficient Self-supervised Learning with data2vec</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-HJOhbT5v3h tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-HJOhbT5v3h .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-HJOhbT5v3h .tatsu-icon_card-title, .tatsu-HJOhbT5v3h .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-HJOhbT5v3h .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-HJOhbT5v3h.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Eli Chen , CTO and Co-Founder | Credo.AI </a></div><div class = "tatsu-icon_card-caption body"><p>Talk: <span data-sheets-value="{" data-sheets-userformat="{">Fine-tuning LLMs on Slack Messages</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-HJK3b6cv23 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-HJK3b6cv23 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-HJK3b6cv23 .tatsu-icon_card-title, .tatsu-HJK3b6cv23 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-HJK3b6cv23 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-HJK3b6cv23.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Michelle Yi - Board Member | Women in Data </a></div><div class = "tatsu-icon_card-caption body"><p>Talk: <span data-sheets-value="{" data-sheets-userformat="{">Building Generative AI Applications: An LLM Case Study</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-hgwnei17bnf54xia tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-hgwnei17bnf54xia .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-hgwnei17bnf54xia .tatsu-icon_card-title, .tatsu-hgwnei17bnf54xia .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-hgwnei17bnf54xia .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-hgwnei17bnf54xia.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Neel Kovelamudi - Software Engineer on Keras Team | Google </a></div><div class = "tatsu-icon_card-caption body"><p>Talk: <span data-sheets-value="{" data-sheets-userformat="{">Keras Core: Keras for TensorFlow, JAX, and PyTorch</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-rJfDuAwn3 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-rJfDuAwn3 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-rJfDuAwn3 .tatsu-icon_card-title, .tatsu-rJfDuAwn3 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-rJfDuAwn3 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-rJfDuAwn3.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Sarah Kefayati - Associate Principal Data Scientist | IBM </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{" data-sheets-validation-definition="{" data-sheets-validation-id="0">Business Talk: Driving Success for Sellers by Infusing AI in CRM Platform</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-HkckY0v3n tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-HkckY0v3n .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-HkckY0v3n .tatsu-icon_card-title, .tatsu-HkckY0v3n .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-HkckY0v3n .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-HkckY0v3n.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Chuying Ma - Senior Data Scientist | Walmart </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{" data-sheets-validation-definition="{" data-sheets-validation-id="0">Talk: A Semi-Supervised Anomaly Detection System Through Ensemble Stacking Algorithm</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-r1tEKAvh3 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-r1tEKAvh3 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-r1tEKAvh3 .tatsu-icon_card-title, .tatsu-r1tEKAvh3 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-r1tEKAvh3 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-r1tEKAvh3.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Alex Liu, Ph.D. - Founder and Director | RMDS Lab </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{" data-sheets-validation-definition="{" data-sheets-validation-id="0">Talk: Accelerating Knowledge Discovery with AI for Enhanced Quality</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-SJy5YCvnn tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-SJy5YCvnn .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-SJy5YCvnn .tatsu-icon_card-title, .tatsu-SJy5YCvnn .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-SJy5YCvnn .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-SJy5YCvnn.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Evie Fowler - Senior Data Scientist | Fulcrum Analytics </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{" data-sheets-validation-definition="{" data-sheets-validation-id="0">Talk: Bridging the Interpretability Gap in Customer Segmentation</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-HJpBEH9hn tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-HJpBEH9hn .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-HJpBEH9hn .tatsu-icon_card-title, .tatsu-HJpBEH9hn .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-HJpBEH9hn .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-HJpBEH9hn.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Walid S. Saba - Senior Research Scientist | Institute for Experiential AI at Northeastern University </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{" data-sheets-validation-definition="{" data-sheets-validation-id="0">Talk: Towards Explainable and Language-Agnostic LLMs</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-S1b3cCP2n tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-S1b3cCP2n .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-S1b3cCP2n .tatsu-icon_card-title, .tatsu-S1b3cCP2n .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-S1b3cCP2n .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-S1b3cCP2n.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Emmanuel Turlay - Founder/CEO | Sematic </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{" data-sheets-validation-definition="{" data-sheets-validation-id="0">Talk: Battle Scars from the MLOps Trenches of the Robotaxi Industry</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-HyJ-sCw2h tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-HyJ-sCw2h .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-HyJ-sCw2h .tatsu-icon_card-title, .tatsu-HyJ-sCw2h .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-HyJ-sCw2h .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-HyJ-sCw2h.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Nirmal Budhathoki - Senior Data Scientist | Microsoft </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{" data-sheets-validation-definition="{" data-sheets-validation-id="0">Talk: Scope of LLMs and GPT Models in Security Domain</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-hgyuv8b4sec5cv4l tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-hgyuv8b4sec5cv4l .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-hgyuv8b4sec5cv4l .tatsu-icon_card-title, .tatsu-hgyuv8b4sec5cv4l .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-hgyuv8b4sec5cv4l .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-hgyuv8b4sec5cv4l.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Robert Osazuwa Ness, PhD - Senior Researcher | Microsoft </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{" data-sheets-validation-definition="{" data-sheets-validation-id="0">Talk: Causality and LLMs</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-HyB5Me0R2 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-HyB5Me0R2 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-HyB5Me0R2 .tatsu-icon_card-title, .tatsu-HyB5Me0R2 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-HyB5Me0R2 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-HyB5Me0R2.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Alessandro Romano - Senior Data Scientist | Kuehne+Nagel </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{" data-sheets-validation-definition="{" data-sheets-validation-id="0">Talk: The Crucial Role of Digital Experimentation and A/B Testing in the AI Landscape</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-r12-7l0Cn tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-r12-7l0Cn .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-r12-7l0Cn .tatsu-icon_card-title, .tatsu-r12-7l0Cn .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-r12-7l0Cn .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-r12-7l0Cn.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Nils Reimers - Director of Machine Learning | Cohere.ai </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{" data-sheets-validation-definition="{" data-sheets-validation-id="0">Talk: Connecting Large Language Models – Common Pitfalls & Challenges</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-Sk7wQeAA3 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-Sk7wQeAA3 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-Sk7wQeAA3 .tatsu-icon_card-title, .tatsu-Sk7wQeAA3 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-Sk7wQeAA3 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-Sk7wQeAA3.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Katie Kuzin - AI Product Manager | Kensho Technologies </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{" data-sheets-validation-definition="{" data-sheets-validation-id="0">Talk: Finance Audio and Automated Speech Recognition – The Perfect Marriage</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-BJjFSeCAh tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-BJjFSeCAh .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-BJjFSeCAh .tatsu-icon_card-title, .tatsu-BJjFSeCAh .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-BJjFSeCAh .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-BJjFSeCAh.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Panos Alexopoulos - Head of Ontology | Textkernel BV </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{" data-sheets-validation-definition="{" data-sheets-validation-id="0">Talk: The Devil in the Details: How Defining an NLP Task can Undermine or Catalyze its Successful Implementation</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-rkjy8eRC2 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-rkjy8eRC2 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-rkjy8eRC2 .tatsu-icon_card-title, .tatsu-rkjy8eRC2 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-rkjy8eRC2 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-rkjy8eRC2.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Jake Bengtson - Principal Technical Evangelist | Cloudera </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{" data-sheets-validation-definition="{" data-sheets-validation-id="0">Talk: How to Deliver Contextually Accurate LLMs</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-rkIN8eCCn tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-rkIN8eCCn .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-rkIN8eCCn .tatsu-icon_card-title, .tatsu-rkIN8eCCn .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-rkIN8eCCn .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-rkIN8eCCn.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Sandy Ryza - Lead Engineer on the Dagster Project | Dagster Labs </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{" data-sheets-validation-definition="{" data-sheets-validation-id="0">Talk: Running Data Quality Checks in Your Data Pipelines</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-HJdnLe0C3 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-HJdnLe0C3 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-HJdnLe0C3 .tatsu-icon_card-title, .tatsu-HJdnLe0C3 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-HJdnLe0C3 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-HJdnLe0C3.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Raghav Bali - Staff Data Scientist; Vishal Natani - Manager, Data Science | Delivery Hero </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{" data-sheets-validation-definition="{" data-sheets-validation-id="0">Talk: Building Robust and Scalable Recommendation Engines for Online Food Delivery</span></p></div></div></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div><div class="tatsu-overlay tatsu-column-overlay tatsu-animate-none" ></div></div><style>.tatsu-row > .tatsu-Bk4hZp5P23.tatsu-column{width: 50%;}.tatsu-Bk4hZp5P23.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: normal;}.tatsu-Bk4hZp5P23 > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-Bk4hZp5P23 > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-Bk4hZp5P23 > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-Bk4hZp5P23 > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width:1377px) {.tatsu-row > .tatsu-Bk4hZp5P23.tatsu-column{width: 50%;}}@media only screen and (min-width:768px) and (max-width: 1024px) {.tatsu-row > .tatsu-Bk4hZp5P23.tatsu-column{width: 50%;}}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-Bk4hZp5P23.tatsu-column{width: 100%;}}</style></div><div class="tatsu-column tatsu-bg-overlay tatsu-one-half tatsu-column-image-none tatsu-column-effect-none tatsu-S193Za5whh" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class = "tatsu-module tatsu-icon_card tatsu-Hki3W6cP22 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-Hki3W6cP22 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-Hki3W6cP22 .tatsu-icon_card-title, .tatsu-Hki3W6cP22 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-Hki3W6cP22 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-Hki3W6cP22.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Peter Norvig - Engineering Director | Education Fellow Google | Stanford Institute for Human-Centered Artificial Intelligence (HAI) </a></div><div class = "tatsu-icon_card-caption body"><p>Keynote: <span data-sheets-value="{" data-sheets-userformat="{">Human-Centered AI</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-B1kyfxRC2 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-B1kyfxRC2 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-B1kyfxRC2 .tatsu-icon_card-title, .tatsu-B1kyfxRC2 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-B1kyfxRC2 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-B1kyfxRC2.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Jack McCauley - Board Trustee at University of California, Berkeley, Former co-founder and Engineer, Oculus VR </a></div><div class = "tatsu-icon_card-caption body"><p>Talk: <span data-sheets-value="{" data-sheets-userformat="{">AI and Video Games : The Evolution</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-HJ33WT5Phn tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-HJ33WT5Phn .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-HJ33WT5Phn .tatsu-icon_card-title, .tatsu-HJ33WT5Phn .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-HJ33WT5Phn .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-HJ33WT5Phn.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Dr. Petar Veličković - Staff Research Scientist | Affiliated Lecturer DeepMind | University of Cambridge </a></div><div class = "tatsu-icon_card-caption body"><p>Talk: <span data-sheets-value="{" data-sheets-userformat="{">Representation Learning on Graphs and Networks</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-rJ63-a5wnn tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-rJ63-a5wnn .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-rJ63-a5wnn .tatsu-icon_card-title, .tatsu-rJ63-a5wnn .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-rJ63-a5wnn .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-rJ63-a5wnn.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Stefanie Molin - Software Engineer, Data Scientist, Chief Information Security Office, Author of Hands-On Data Analysis with Pandas | Bloombergv </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Talk: Data Morph: A Cautionary Tale of Summary Statistics</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-rkCn-a9w2n tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-rkCn-a9w2n .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-rkCn-a9w2n .tatsu-icon_card-title, .tatsu-rkCn-a9w2n .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-rkCn-a9w2n .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-rkCn-a9w2n.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Angad Arora - Manufacturing Data Scientist Google </a></div><div class = "tatsu-icon_card-caption body"><p>Talk: <span data-sheets-value="{" data-sheets-userformat="{">Data Science Applied to Manufacturing Problems</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-SJyg2-a9w2h tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-SJyg2-a9w2h .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-SJyg2-a9w2h .tatsu-icon_card-title, .tatsu-SJyg2-a9w2h .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-SJyg2-a9w2h .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-SJyg2-a9w2h.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Dominic Bohan - Co-Founder | StoryIQ </a></div><div class = "tatsu-icon_card-caption body"><p>Talk: <span data-sheets-value="{" data-driven="" data-sheets-userformat="{">Building a Data-Driven Workforce</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-rygpdRDhn tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-rygpdRDhn .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-rygpdRDhn .tatsu-icon_card-title, .tatsu-rygpdRDhn .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-rygpdRDhn .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-rygpdRDhn.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Joep Kokkeler - Senior Data Engineer | Dataworkz NL </a></div><div class = "tatsu-icon_card-caption body"><p>Talk: <span data-sheets-value="{" data-sheets-userformat="{">Capturing CAP in a Kappa Data Architecture</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-ry-ztCvnh tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-ry-ztCvnh .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-ry-ztCvnh .tatsu-icon_card-title, .tatsu-ry-ztCvnh .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-ry-ztCvnh .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-ry-ztCvnh.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Doris Lee - CEO and Cofounder | Ponder </a></div><div class = "tatsu-icon_card-caption body"><p>Talk: <span data-sheets-value="{" data-sheets-userformat="{">Scaling your Data Science Workflows by Changing a Single Line of Code</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-SyuPtAP3n tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-SyuPtAP3n .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-SyuPtAP3n .tatsu-icon_card-title, .tatsu-SyuPtAP3n .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-SyuPtAP3n .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-SyuPtAP3n.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Amber Roberts - Data Scientist, Growth Lead | Arize AI </a></div><div class = "tatsu-icon_card-caption body"><p>Talk: <span data-sheets-value="{" data-sheets-userformat="{">Troubleshooting and Measuring Embedding/Vector Drift for Production Deployments of Language Models</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-Hyo2FAPn2 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-Hyo2FAPn2 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-Hyo2FAPn2 .tatsu-icon_card-title, .tatsu-Hyo2FAPn2 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-Hyo2FAPn2 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-Hyo2FAPn2.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Kabir Nagrecha - PhD Student | UC San Diego </a></div><div class = "tatsu-icon_card-caption body"><p>Talk: <span data-sheets-value="{" data-sheets-userformat="{">Democratizing Fine-tuning of Open-Source Large Models with Joint Systems Optimization</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-HJmVcAw23 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-HJmVcAw23 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-HJmVcAw23 .tatsu-icon_card-title, .tatsu-HJmVcAw23 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-HJmVcAw23 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-HJmVcAw23.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Sanjay Jinturkar - Senior Director, MySQL HeatWave | Oracle, Sandeep Agrawal, PhD - Consulting Principal Member of Technical Staff | Oracle </a></div><div class = "tatsu-icon_card-caption body"><p>Talk: <span data-sheets-value="{" data-sheets-userformat="{">A Unified and User Friendly Approach to Develop ML Solutions in MySQL HeatWave AutoML</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-BkG1s0Phh tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-BkG1s0Phh .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-BkG1s0Phh .tatsu-icon_card-title, .tatsu-BkG1s0Phh .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-BkG1s0Phh .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-BkG1s0Phh.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Anna Jung - Sr. ML Open Source Engineer | VMware </a></div><div class = "tatsu-icon_card-caption body"><p>Talk: <span data-sheets-value="{" data-sheets-userformat="{">Attack on Machine Learning, Defend with MLOps</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-B14SsRvnn tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-B14SsRvnn .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-B14SsRvnn .tatsu-icon_card-title, .tatsu-B14SsRvnn .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-B14SsRvnn .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-B14SsRvnn.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Teodora Sechkova - Open Source Software Engineer | VMware </a></div><div class = "tatsu-icon_card-caption body"><p>Talk: <span data-sheets-value="{" data-sheets-userformat="{">Security First, Create a Robust Machine Learning Model</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-B1iIfeAR2 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-B1iIfeAR2 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-B1iIfeAR2 .tatsu-icon_card-title, .tatsu-B1iIfeAR2 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-B1iIfeAR2 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-B1iIfeAR2.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Robert Crowe - Product Manager, MLOps and TF OSS | Google </a></div><div class = "tatsu-icon_card-caption body"><p>Talk: <span data-sheets-value="{" data-sheets-userformat="{">MLOps v LMOps – What’s Different?</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-rJ9pfl0R3 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-rJ9pfl0R3 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-rJ9pfl0R3 .tatsu-icon_card-title, .tatsu-rJ9pfl0R3 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-rJ9pfl0R3 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-rJ9pfl0R3.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Patrick Hall - Assistant professor | Principal Scientist George Washington University School of Business| BNH.AI </a></div><div class = "tatsu-icon_card-caption body"><p>Talk: <span data-sheets-value="{" data-sheets-userformat="{">Adopting Language Models Requires Risk Management — This is How</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-B1YNXlCCh tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-B1YNXlCCh .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-B1YNXlCCh .tatsu-icon_card-title, .tatsu-B1YNXlCCh .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-B1YNXlCCh .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-B1YNXlCCh.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > David Mertz, Ph.D. - Director of Epistemology | KDM Training </a></div><div class = "tatsu-icon_card-caption body"><p>Talk: <span data-sheets-value="{" data-sheets-userformat="{">Attribution and Moral Rights in Generative AI</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-S1Jc7l0Cn tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-S1Jc7l0Cn .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-S1Jc7l0Cn .tatsu-icon_card-title, .tatsu-S1Jc7l0Cn .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-S1Jc7l0Cn .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-S1Jc7l0Cn.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Denny Lee - Sr. Staff Developer Advocate | Databricks </a></div><div class = "tatsu-icon_card-caption body"><p>Talk: <span data-sheets-value="{" data-sheets-userformat="{">The English SDK for Apache Spark™</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-Hyll4eC03 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-Hyll4eC03 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-Hyll4eC03 .tatsu-icon_card-title, .tatsu-Hyll4eC03 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-Hyll4eC03 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-Hyll4eC03.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Sergey Yurgenson - Head of Semantic Data Science | Featurebyte </a></div><div class = "tatsu-icon_card-caption body"><p>Talk: <span data-sheets-value="{" data-sheets-userformat="{">Integrating Language Models for Automating Feature Engineering Ideation</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-Hy8hrxR02 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-Hy8hrxR02 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-Hy8hrxR02 .tatsu-icon_card-title, .tatsu-Hy8hrxR02 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-Hy8hrxR02 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-Hy8hrxR02.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Isha Ghodgaonkar - Machine Learning Developer Advocate | Hewlett Packard Enterprise (HPE) </a></div><div class = "tatsu-icon_card-caption body"><p>Talk: <span data-sheets-value="{" data-sheets-userformat="{">Machine Learning Development Environment and the Open Source ML Advantage</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-S1hUUgCC3 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-S1hUUgCC3 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-S1hUUgCC3 .tatsu-icon_card-title, .tatsu-S1hUUgCC3 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-S1hUUgCC3 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-S1hUUgCC3.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Hudson Buzby - Solution Architect | Qwak </a></div><div class = "tatsu-icon_card-caption body"><p>Talk: <span data-sheets-value="{" data-sheets-userformat="{">From Raw Data through Vectors to a Comprehensive Recommendation Model</span></p></div></div></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div><div class="tatsu-overlay tatsu-column-overlay tatsu-animate-none" ></div></div><style>.tatsu-row > .tatsu-S193Za5whh.tatsu-column{width: 50%;}.tatsu-S193Za5whh.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: normal;}.tatsu-S193Za5whh > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-S193Za5whh > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-S193Za5whh > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-S193Za5whh > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width:1377px) {.tatsu-row > .tatsu-S193Za5whh.tatsu-column{width: 50%;}}@media only screen and (min-width:768px) and (max-width: 1024px) {.tatsu-row > .tatsu-S193Za5whh.tatsu-column{width: 50%;}}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-S193Za5whh.tatsu-column{width: 100%;}}</style></div></div></div><div class="tatsu-row-wrap tatsu-wrap tatsu-row-one-col tatsu-row-has-one-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-Byxg2-65wh3" ><div class="tatsu-row " ><div class="tatsu-column tatsu-bg-overlay tatsu-column-empty tatsu-one-col tatsu-column-image-none tatsu-column-effect-none tatsu-S1Wg2W6cv2h" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div><div class="tatsu-overlay tatsu-column-overlay tatsu-animate-none" ></div></div><style>.tatsu-row > .tatsu-S1Wg2W6cv2h.tatsu-column{width: 100%;}.tatsu-S1Wg2W6cv2h.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: normal;}.tatsu-S1Wg2W6cv2h > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-S1Wg2W6cv2h > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-S1Wg2W6cv2h > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-S1Wg2W6cv2h > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width:1377px) {.tatsu-row > .tatsu-S1Wg2W6cv2h.tatsu-column{width: 100%;}}@media only screen and (min-width:768px) and (max-width: 1024px) {.tatsu-row > .tatsu-S1Wg2W6cv2h.tatsu-column{width: 100%;}}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-S1Wg2W6cv2h.tatsu-column{width: 100%;}}</style></div></div></div></div><div class="tatsu-section-background-wrap"><div class = "tatsu-section-background" ></div></div><div class="tatsu-overlay tatsu-section-overlay"></div><style>.tatsu-Bkn-6cP23.tatsu-section{background-color: rgba(0,0,0,1);}.tatsu-Bkn-6cP23 .tatsu-section-pad{padding: 90px 0px 90px 0px;}.tatsu-Bkn-6cP23 .tatsu-section-offset-wrap{transform: translateY(-0px);}.tatsu-Bkn-6cP23 > .tatsu-bottom-divider{z-index: 9999;}.tatsu-Bkn-6cP23 > .tatsu-top-divider{z-index: 9999;}.tatsu-Bkn-6cP23 .tatsu-section-overlay{mix-blend-mode: normal;}</style></div><div class="tatsu-hgwjj0vxfechsbn9 tatsu-section tatsu-bg-overlay tatsu-hide-mobile tatsu-hide-tablet tatsu-hide-laptop tatsu-hide-desktop tatsu-clearfix" data-title="Training& Workshops" data-headerscheme="background--dark"><div class='tatsu-section-pad clearfix' data-padding='{"d":"90px 0px 90px 0px"}' data-padding-top='90px'><div class="tatsu-row-wrap tatsu-wrap tatsu-row-one-col tatsu-row-has-one-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-hgwjj0vxi5fn45s5" ><div class="tatsu-row " ><div class="tatsu-column tatsu-bg-overlay tatsu-one-col tatsu-column-image-none tatsu-column-effect-none tatsu-hgwjj0vxla4pu8e5" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class="tatsu-module tatsu-text-block-wrap tatsu-hgwjj0vxn46y7fma "><div class="tatsu-text-inner tatsu-align-center clearfix" ><style>.tatsu-hgwjj0vxn46y7fma.tatsu-text-block-wrap .tatsu-text-inner{width: 100%;text-align: left;background-color: rgba(0,0,0,1);}</style><p style="text-align: center;"><span style="color: #ffffff; font-size: 24pt;">Past Training & Workshops</span></p></div></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div><div class="tatsu-overlay tatsu-column-overlay tatsu-animate-none" ></div></div><style>.tatsu-row > .tatsu-hgwjj0vxla4pu8e5.tatsu-column{width: 100%;}.tatsu-hgwjj0vxla4pu8e5.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: normal;}.tatsu-hgwjj0vxla4pu8e5 > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-hgwjj0vxla4pu8e5 > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-hgwjj0vxla4pu8e5 > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-hgwjj0vxla4pu8e5 > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width:1377px) {.tatsu-row > .tatsu-hgwjj0vxla4pu8e5.tatsu-column{width: 100%;}}@media only screen and (min-width:768px) and (max-width: 1024px) {.tatsu-row > .tatsu-hgwjj0vxla4pu8e5.tatsu-column{width: 100%;}}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-hgwjj0vxla4pu8e5.tatsu-column{width: 100%;}}</style></div></div></div><div class="tatsu-row-wrap tatsu-wrap tatsu-row-has-one-half tatsu-row-has-two-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-hgwjj0vxondmicpj" ><div class="tatsu-row " ><div class="tatsu-column tatsu-bg-overlay tatsu-one-half tatsu-column-image-none tatsu-column-effect-none tatsu-hgwjj0vxr8ccncpz" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class = "tatsu-module tatsu-icon_card tatsu-hgwjj0vxtb3ga21t tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-hgwjj0vxtb3ga21t .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-hgwjj0vxtb3ga21t .tatsu-icon_card-title, .tatsu-hgwjj0vxtb3ga21t .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-hgwjj0vxtb3ga21t .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-hgwjj0vxtb3ga21t.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Oliver Zeigermann - Machine Learning Architect | Freelancer </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Training: MLOps: Monitoring and Managing Drift</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-hgwjj0vxucfbjuc tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-hgwjj0vxucfbjuc .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-hgwjj0vxucfbjuc .tatsu-icon_card-title, .tatsu-hgwjj0vxucfbjuc .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-hgwjj0vxucfbjuc .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-hgwjj0vxucfbjuc.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Rajiv Shah, PhD - Machine Learning Engineer | Hugging Face </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Tutorial: Evaluation Techniques for Large Language Models</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-hgwjj0vxv7g4l0j8 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-hgwjj0vxv7g4l0j8 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-hgwjj0vxv7g4l0j8 .tatsu-icon_card-title, .tatsu-hgwjj0vxv7g4l0j8 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-hgwjj0vxv7g4l0j8 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-hgwjj0vxv7g4l0j8.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Mark Saroufim - Engineer on PyTorch | Meta </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Tutorial: Machine Learning Has Become Necromancy</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-hgwjj0vxw0a57ypv tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-hgwjj0vxw0a57ypv .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-hgwjj0vxw0a57ypv .tatsu-icon_card-title, .tatsu-hgwjj0vxw0a57ypv .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-hgwjj0vxw0a57ypv .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-hgwjj0vxw0a57ypv.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Brian Lucena - Principal | Numeristical </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Training: Uncertainty Quantification: Approaches and Methods</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-hgwjj0vxwq8sytng tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-hgwjj0vxwq8sytng .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-hgwjj0vxwq8sytng .tatsu-icon_card-title, .tatsu-hgwjj0vxwq8sytng .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-hgwjj0vxwq8sytng .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-hgwjj0vxwq8sytng.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Martin Musiol - Co-Founder and Instructor | Principal Data Science Manager Generative AI.net | Infosys Consulting </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Training: Generative AI, Autonomous AI Agents, and AGI – How new Advancements in AI will Improve the Products we Build</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-hgwm6ugsvk9zg8bm tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-hgwm6ugsvk9zg8bm .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-hgwm6ugsvk9zg8bm .tatsu-icon_card-title, .tatsu-hgwm6ugsvk9zg8bm .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-hgwm6ugsvk9zg8bm .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-hgwm6ugsvk9zg8bm.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Jim Dowling - CEO | Hopsworks </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Workshop: Personalizing LLMs with a Feature Store</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-hgwm9et6of5epm4e tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-hgwm9et6of5epm4e .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-hgwm9et6of5epm4e .tatsu-icon_card-title, .tatsu-hgwm9et6of5epm4e .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-hgwm9et6of5epm4e .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-hgwm9et6of5epm4e.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Mike Taylor - Owner | Saxifrage </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Tutorial: Prompt Optimization with GPT-4 and Langchain</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-hgwmaqqkg4est947 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-hgwmaqqkg4est947 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-hgwmaqqkg4est947 .tatsu-icon_card-title, .tatsu-hgwmaqqkg4est947 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-hgwmaqqkg4est947 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-hgwmaqqkg4est947.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Jonas Mueller - Chief Scientist and Co-Founder | Cleanlab </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-centric="" data-sheets-userformat="{">Tutorial: How to Practice Data-Centric AI and Have AI improve its Own Dataset</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-hgwmf15q0zgjbnvd tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-hgwmf15q0zgjbnvd .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-hgwmf15q0zgjbnvd .tatsu-icon_card-title, .tatsu-hgwmf15q0zgjbnvd .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-hgwmf15q0zgjbnvd .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-hgwmf15q0zgjbnvd.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Alison Cossette - Developer Advocate - Data Science | Neo4j </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Workshop: Bridging the Gap: Light Code Solutions to Uniting Social Science and Modern Knowledge Graphs</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-hgyv3v6ivd8wapfy tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-hgyv3v6ivd8wapfy .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-hgyv3v6ivd8wapfy .tatsu-icon_card-title, .tatsu-hgyv3v6ivd8wapfy .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-hgyv3v6ivd8wapfy .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-hgyv3v6ivd8wapfy.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Gwendolyn D. Stripling, PhD - Lead AI & ML Content Developer | Google Cloud </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Tutorial: No-Code and Low-Code AI: A Practical Project Driven Approach to ML</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-hh06sa4dnq9muaxx tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-hh06sa4dnq9muaxx .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-hh06sa4dnq9muaxx .tatsu-icon_card-title, .tatsu-hh06sa4dnq9muaxx .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-hh06sa4dnq9muaxx .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-hh06sa4dnq9muaxx.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Serg Masis - Lead Data Scientist, Best Selling Author of AI/ML books | Syngenta </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Workshop: Facial Recognition from Scratch with Python and JS</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-SJCWRkAAn tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-SJCWRkAAn .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-SJCWRkAAn .tatsu-icon_card-title, .tatsu-SJCWRkAAn .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-SJCWRkAAn .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-SJCWRkAAn.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Matt Harrison - Python & Data Science Corporate Trainer | Consultant | MetaSnake </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Workshop: Machine Learning with XGBoost</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-Sk4E0J0A3 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-Sk4E0J0A3 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-Sk4E0J0A3 .tatsu-icon_card-title, .tatsu-Sk4E0J0A3 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-Sk4E0J0A3 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-Sk4E0J0A3.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Matt Harrison - Python & Data Science Corporate Trainer | Consultant | MetaSnake </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Workshop: Idiomatic Pandas</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-rJ-JyeRAh tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-rJ-JyeRAh .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-rJ-JyeRAh .tatsu-icon_card-title, .tatsu-rJ-JyeRAh .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-rJ-JyeRAh .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-rJ-JyeRAh.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Shashank Prasanna - AI Developer Advocate | Modular </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Workshop: Overview of Mojo🔥: Usability of Python, Performance of C</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-SyADkeCC2 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-SyADkeCC2 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-SyADkeCC2 .tatsu-icon_card-title, .tatsu-SyADkeCC2 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-SyADkeCC2 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-SyADkeCC2.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Wes Madrigal - ML Engineer | Mad Consulting </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Workshop: Using Graphs for Large Feature Engineering Pipelines</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-rkzgxeARn tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-rkzgxeARn .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-rkzgxeARn .tatsu-icon_card-title, .tatsu-rkzgxeARn .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-rkzgxeARn .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-rkzgxeARn.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Bob Foreman - Software Engineering Lead | LexisNexis Risk Solutions </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Workshop: Data for Social Good – Find Your Paradise!</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-ByxOggAR3 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-ByxOggAR3 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-ByxOggAR3 .tatsu-icon_card-title, .tatsu-ByxOggAR3 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-ByxOggAR3 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-ByxOggAR3.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Philip Wauters - Customer Success Manager and Value Engineer | Tangent Works </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Workshop: Learn how to Efficiently Build and Operationalize Time Series Models in 2023</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-BylAelCCh tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-BylAelCCh .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-BylAelCCh .tatsu-icon_card-title, .tatsu-BylAelCCh .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-BylAelCCh .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-BylAelCCh.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Krishnaram Kenthapadi - Chief AI Officer & Chief Scientist | Fiddler AI </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Tutorial: Deploying Trustworthy Generative AI</span></p></div></div></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div><div class="tatsu-overlay tatsu-column-overlay tatsu-animate-none" ></div></div><style>.tatsu-row > .tatsu-hgwjj0vxr8ccncpz.tatsu-column{width: 50%;}.tatsu-hgwjj0vxr8ccncpz.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: normal;}.tatsu-hgwjj0vxr8ccncpz > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-hgwjj0vxr8ccncpz > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-hgwjj0vxr8ccncpz > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-hgwjj0vxr8ccncpz > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width:1377px) {.tatsu-row > .tatsu-hgwjj0vxr8ccncpz.tatsu-column{width: 50%;}}@media only screen and (min-width:768px) and (max-width: 1024px) {.tatsu-row > .tatsu-hgwjj0vxr8ccncpz.tatsu-column{width: 50%;}}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-hgwjj0vxr8ccncpz.tatsu-column{width: 100%;}}</style></div><div class="tatsu-column tatsu-bg-overlay tatsu-one-half tatsu-column-image-none tatsu-column-effect-none tatsu-hgwjj0vxxx4kpmg1" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class = "tatsu-module tatsu-icon_card tatsu-hgwjj0vxzt5aj70w tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-hgwjj0vxzt5aj70w .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-hgwjj0vxzt5aj70w .tatsu-icon_card-title, .tatsu-hgwjj0vxzt5aj70w .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-hgwjj0vxzt5aj70w .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-hgwjj0vxzt5aj70w.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Sinan Ozdemir - AI & LLM Expert | Author | Founder + CTO | LoopGenius </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Workshop: Aligning Open-source LLMs Using Reinforcement Learning from Feedback</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-hgwjj0vy0k9lpys0 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-hgwjj0vy0k9lpys0 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-hgwjj0vy0k9lpys0 .tatsu-icon_card-title, .tatsu-hgwjj0vy0k9lpys0 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-hgwjj0vy0k9lpys0 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-hgwjj0vy0k9lpys0.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Jerry Liu - Co-founder and CEO | LlamaIndex </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Workshop: Building LLM-powered Knowledge Workers over Your Data with LlamaIndex</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-hgwjj0vy1ag3wp4o tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-hgwjj0vy1ag3wp4o .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-hgwjj0vy1ag3wp4o .tatsu-icon_card-title, .tatsu-hgwjj0vy1ag3wp4o .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-hgwjj0vy1ag3wp4o .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-hgwjj0vy1ag3wp4o.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Sandeep Singh - Head of Applied AI | Computer Vision | Beans.ai </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Workshop: Stable Diffusion: A New Frontier for Text-to-Image Paradigm</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-hgwjj0vy2029lb09 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-hgwjj0vy2029lb09 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-hgwjj0vy2029lb09 .tatsu-icon_card-title, .tatsu-hgwjj0vy2029lb09 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-hgwjj0vy2029lb09 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-hgwjj0vy2029lb09.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > James Phoenix - CTO | Vexpower </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Tutorial: Automating Business Processes Using LangChain</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-hgwjj0vy2r9k8ra tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-hgwjj0vy2r9k8ra .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-hgwjj0vy2r9k8ra .tatsu-icon_card-title, .tatsu-hgwjj0vy2r9k8ra .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-hgwjj0vy2r9k8ra .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-hgwjj0vy2r9k8ra.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Parul Pandey , Principal Data Scientist | H2O.ai </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Tutorial: Machine Learning for High-Risk Applications – Techniques for Responsible AIO</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-hgwm6ymu4d1815ss tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-hgwm6ymu4d1815ss .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-hgwm6ymu4d1815ss .tatsu-icon_card-title, .tatsu-hgwm6ymu4d1815ss .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-hgwm6ymu4d1815ss .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-hgwm6ymu4d1815ss.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Jeff Tao - Founder & CEO | TDengine </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Workshop: What is a Time-series Database and Why do I Need One?</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-hgwm9id6q2o4poo tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-hgwm9id6q2o4poo .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-hgwm9id6q2o4poo .tatsu-icon_card-title, .tatsu-hgwm9id6q2o4poo .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-hgwm9id6q2o4poo .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-hgwm9id6q2o4poo.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Dr. Andre Franca - VP of Research and Development | causaLens </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Workshop: Causal AI: from Data to Action</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-hgwmauxtvp8u2i5k tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-hgwmauxtvp8u2i5k .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-hgwmauxtvp8u2i5k .tatsu-icon_card-title, .tatsu-hgwmauxtvp8u2i5k .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-hgwmauxtvp8u2i5k .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-hgwmauxtvp8u2i5k.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Geeta Shankar - Software Engineer | Salesforce </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Workshop: Anomaly Detection for CRM Production Data</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-hgwmf61heqf44esd tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-hgwmf61heqf44esd .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-hgwmf61heqf44esd .tatsu-icon_card-title, .tatsu-hgwmf61heqf44esd .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-hgwmf61heqf44esd .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-hgwmf61heqf44esd.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Suhas Pai - Chief Technology Officer | Bedrock AI </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Workshop: Beyond Demos and Prototypes: How to Build Production-Ready Applications Using Open-Source LLMs</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-hgyv41j8shahe1t0 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-hgyv41j8shahe1t0 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-hgyv41j8shahe1t0 .tatsu-icon_card-title, .tatsu-hgyv41j8shahe1t0 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-hgyv41j8shahe1t0 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-hgyv41j8shahe1t0.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Amy Hodler - Founder, Consultant | GraphGeeks.org and Michelle Yi - Board Member | Women In Data </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Workshop: Graphs: The Next Frontier of GenAI Explainability</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-SJFxa1C0h tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-SJFxa1C0h .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-SJFxa1C0h .tatsu-icon_card-title, .tatsu-SJFxa1C0h .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-SJFxa1C0h .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-SJFxa1C0h.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Ramon Perez -Developer Advocate | Instructor | Seldon | Decoded </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Half-day Training: Architecting Data: A Deep Dive Into the World of Synthetic Data</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-HkhtpJAAh tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-HkhtpJAAh .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-HkhtpJAAh .tatsu-icon_card-title, .tatsu-HkhtpJAAh .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-HkhtpJAAh .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-HkhtpJAAh.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Greg Loug hnane - Lead Instructor, Building with LLMs | FourthBrain; Chris Alexiuk - Head of LLMs | AI Makerspace </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Half-day Training: Retrieval Augmented Generation (RAG) 101: Building an Open-Source “ChatGPT for Your Data” with Llama 2, LangChain, and Pinecone</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-rJuOAkRR2 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-rJuOAkRR2 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-rJuOAkRR2 .tatsu-icon_card-title, .tatsu-rJuOAkRR2 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-rJuOAkRR2 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-rJuOAkRR2.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Vincent Granville - CEO and Executive Machine Learning Scientist | MLtechniques.com </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Tutorial: Massively Speed-Up your Learning Algorithm, with Stochastic Thinning</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-HJYVyeRRn tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-HJYVyeRRn .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-HJYVyeRRn .tatsu-icon_card-title, .tatsu-HJYVyeRRn .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-HJYVyeRRn .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-HJYVyeRRn.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Fabiana Clemente - Co-founder and CDO | YData </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Workshop: Missing Data: A Synthetic Data Approach for Missing Data Imputation</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-S1331gRAh tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-S1331gRAh .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-S1331gRAh .tatsu-icon_card-title, .tatsu-S1331gRAh .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-S1331gRAh .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-S1331gRAh.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Andrew Dai - Principal Software Engineer | Google </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Tutorial: A Background to LLMs and Intro to PaLM 2: A Smaller, Faster and More Capable LLM</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-Hkw7glCR3 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-Hkw7glCR3 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-Hkw7glCR3 .tatsu-icon_card-title, .tatsu-Hkw7glCR3 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-Hkw7glCR3 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-Hkw7glCR3.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Valentina Alto - Azure Specialist - Data and Artificial Intelligence | Microsoft </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Workshop: The AI Paradigm Shift: Under the Hood of a Large Language Models</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-BJicgeRA3 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-BJicgeRA3 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-BJicgeRA3 .tatsu-icon_card-title, .tatsu-BJicgeRA3 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-BJicgeRA3 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-BJicgeRA3.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Vino Duraisamy - Developer Advocate | Snowflake </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Workshop: The Rise of a Full Stack Data Scientist: Powered by Python</span></p></div></div></div><div class = "tatsu-module tatsu-icon_card tatsu-SJuWblAC2 tatsu-icon_card-type-image tatsu-icon_card-style1 tatsu-icon_card-align-left tatsu-icon_card-vertical-align-top tatsu-icon_large" > <style>.tatsu-SJuWblAC2 .tatsu-icon_card-icon{background: url(https://odsc.com/wp-content/uploads/2023/08/Screenshot-2023-08-03-at-13.00.01-Cropped-150x150.png) center scroll no-repeat;background-size: cover;box-shadow: 0px 0px 0px 0px rgba(0,0,0,0);}.tatsu-SJuWblAC2 .tatsu-icon_card-title, .tatsu-SJuWblAC2 .tatsu-icon_card-title a{color: rgba(155,155,155,1) ;}.tatsu-SJuWblAC2 .tatsu-icon_card-caption{color: rgba(255,255,255,1) ;}.tatsu-SJuWblAC2.tatsu-module{margin: 0 0 30px 0;}</style><div class = "tatsu-icon_card-icon tatsu-img-plain"></div><div class = "tatsu-icon_card-title-caption"><div class = "tatsu-icon_card-title h6"> <a href = "#" > Amit Sangani - Director of Partner Engineering | Meta </a></div><div class = "tatsu-icon_card-caption body"><p><span data-sheets-value="{" data-sheets-userformat="{">Workshop: Building Using Llama 2</span></p></div></div></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div><div class="tatsu-overlay tatsu-column-overlay tatsu-animate-none" ></div></div><style>.tatsu-row > .tatsu-hgwjj0vxxx4kpmg1.tatsu-column{width: 50%;}.tatsu-hgwjj0vxxx4kpmg1.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: normal;}.tatsu-hgwjj0vxxx4kpmg1 > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-hgwjj0vxxx4kpmg1 > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-hgwjj0vxxx4kpmg1 > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-hgwjj0vxxx4kpmg1 > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width:1377px) {.tatsu-row > .tatsu-hgwjj0vxxx4kpmg1.tatsu-column{width: 50%;}}@media only screen and (min-width:768px) and (max-width: 1024px) {.tatsu-row > .tatsu-hgwjj0vxxx4kpmg1.tatsu-column{width: 50%;}}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-hgwjj0vxxx4kpmg1.tatsu-column{width: 100%;}}</style></div></div></div><div class="tatsu-row-wrap tatsu-wrap tatsu-row-one-col tatsu-row-has-one-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-hgwjj0vy3ee7w9ad" ><div class="tatsu-row " ><div class="tatsu-column tatsu-bg-overlay tatsu-column-empty tatsu-one-col tatsu-column-image-none tatsu-column-effect-none tatsu-hgwjj0vy5ffubvcc" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div><div class="tatsu-overlay tatsu-column-overlay tatsu-animate-none" ></div></div><style>.tatsu-row > .tatsu-hgwjj0vy5ffubvcc.tatsu-column{width: 100%;}.tatsu-hgwjj0vy5ffubvcc.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: normal;}.tatsu-hgwjj0vy5ffubvcc > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-hgwjj0vy5ffubvcc > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-hgwjj0vy5ffubvcc > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-hgwjj0vy5ffubvcc > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width:1377px) {.tatsu-row > .tatsu-hgwjj0vy5ffubvcc.tatsu-column{width: 100%;}}@media only screen and (min-width:768px) and (max-width: 1024px) {.tatsu-row > .tatsu-hgwjj0vy5ffubvcc.tatsu-column{width: 100%;}}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-hgwjj0vy5ffubvcc.tatsu-column{width: 100%;}}</style></div></div></div></div><div class="tatsu-section-background-wrap"><div class = "tatsu-section-background" ></div></div><div class="tatsu-overlay tatsu-section-overlay"></div><style>.tatsu-hgwjj0vxfechsbn9.tatsu-section{background-color: rgba(0,0,0,1);}.tatsu-hgwjj0vxfechsbn9 .tatsu-section-pad{padding: 90px 0px 90px 0px;}.tatsu-hgwjj0vxfechsbn9 .tatsu-section-offset-wrap{transform: translateY(-0px);}.tatsu-hgwjj0vxfechsbn9 > .tatsu-bottom-divider{z-index: 9999;}.tatsu-hgwjj0vxfechsbn9 > .tatsu-top-divider{z-index: 9999;}.tatsu-hgwjj0vxfechsbn9 .tatsu-section-overlay{mix-blend-mode: normal;}</style></div><div class="tatsu-gso398arlla8257q tatsu-section tatsu-hide-0 tatsu-hide-laptop tatsu-hide-desktop tatsu-hide-tablet tatsu-hide-mobile tatsu-clearfix" data-title="" data-headerscheme="background--dark"><div class='tatsu-section-pad clearfix' data-padding='{"d":"30px 0px 30px 0px"}' data-padding-top='30px'><div class="tatsu-row-wrap tatsu-wrap tatsu-row-one-col tatsu-row-has-one-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-gso398arp1arl9lk" ><div class="tatsu-row " ><div class="tatsu-column tatsu-column-no-bg tatsu-one-col tatsu-column-image-none tatsu-column-effect-none tatsu-gso398artbfr4qws" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class="tatsu-module tatsu-call-to-action tatsu-clearfix tatsu-animate tatsu-gso398arx7doweas " data-animation="fadeIn" ><div class="tatsu-cta-inner"><h3 class="tatsu-action-content" >Register for ODSC West 2023 - Oct 30th - Nov 2nd</h3><a class="mediumbtn tatsu-button rounded tatsu-action-button " href="/california/#register" data-gdpr-atts={} target="_blank"><span>Save 50% NOW</span></a><style>.tatsu-gso398arx7doweas.tatsu-call-to-action{background-color: rgba(16,87,171,1);}.tatsu-gso398arx7doweas .tatsu-action-content{color: rgba(255,255,255,1) ;}.tatsu-gso398arx7doweas .tatsu-action-button{background: rgba(255,186,0,1);border-width: 3px;border-color: rgba(0,0,0,1); }.tatsu-gso398arx7doweas .tatsu-action-button:hover{background: rgba(207,7,0,1);border-color: rgba(0,0,0,1); }.tatsu-gso398arx7doweas .tatsu-action-button span{color: rgba(0,0,0,1) ;}.tatsu-gso398arx7doweas .tatsu-action-button:hover span{color: rgba(255,255,255,1) ;}.tatsu-gso398arx7doweas{margin: 0px 0px 0px 0px;}</style></div></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div></div><style>.tatsu-row > .tatsu-gso398artbfr4qws.tatsu-column{width: 100%;}.tatsu-gso398artbfr4qws.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: none;}.tatsu-gso398artbfr4qws > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-gso398artbfr4qws > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-gso398artbfr4qws > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-gso398artbfr4qws > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}</style></div></div></div></div><div class="tatsu-section-background-wrap"><div class = "tatsu-section-background" ></div></div><style>.tatsu-gso398arlla8257q .tatsu-section-pad{padding: 30px 0px 30px 0px;}.tatsu-gso398arlla8257q > .tatsu-bottom-divider{z-index: 9999;}.tatsu-gso398arlla8257q > .tatsu-top-divider{z-index: 9999;}</style></div><div class="tatsu-gso399v3koe49dnx tatsu-section tatsu-clearfix" data-title="" data-headerscheme="background--dark"><div class='tatsu-section-pad clearfix' data-padding='{"d":"px 0px 90px 0px"}' data-padding-top='px'></div><div class="tatsu-section-background-wrap"><div class = "tatsu-section-background" ></div></div><style>.tatsu-gso399v3koe49dnx .tatsu-section-pad{padding: px 0px 90px 0px;}.tatsu-gso399v3koe49dnx > .tatsu-bottom-divider{z-index: 9999;}.tatsu-gso399v3koe49dnx > .tatsu-top-divider{z-index: 9999;}</style></div><div id="vcworks" class="tatsu-gso4e6n9418nmat9 tatsu-section tatsu-hide-0 tatsu-hide-desktop tatsu-hide-tablet tatsu-hide-mobile tatsu-hide-laptop tatsu-clearfix" data-title="" data-headerscheme="background--dark"><div class='tatsu-section-pad clearfix' data-padding='{"d":"10px 0px 30px 0px "}' data-padding-top='10px'><div class="tatsu-row-wrap tatsu-row-full-width tatsu-row-one-col tatsu-row-has-one-cols tatsu-zero-margin tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-gso399v405abiknm" ><div class="tatsu-row " ><div class="tatsu-column tatsu-column-no-bg tatsu-one-col tatsu-column-image-none tatsu-column-effect-none tatsu-gso399v44yfjatr" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class="tatsu-module tatsu-call-to-action tatsu-clearfix tatsu-gso399v49b92dq0d " ><div class="tatsu-cta-inner"><h5 class="tatsu-action-content" >See our Program Summary for an Event Overview</h5><a class="mediumbtn tatsu-button rounded tatsu-action-button " href="/california/schedule-overview/" data-gdpr-atts={} target="_blank"><span>Program Summary</span></a><style>.tatsu-gso399v49b92dq0d.tatsu-call-to-action{background-color: rgba(99,99,99,1);}.tatsu-gso399v49b92dq0d .tatsu-action-content{color: #ffffff ;}.tatsu-gso399v49b92dq0d .tatsu-action-button{background: rgba(255,255,255,1);border-width: 1px;border-color: rgba(255,0,0,1); }.tatsu-gso399v49b92dq0d .tatsu-action-button:hover{background: rgba(255,0,0,1);border-color: rgba(0,0,0,1); }.tatsu-gso399v49b92dq0d .tatsu-action-button span{color: rgba(255,0,0,1) ;}.tatsu-gso399v49b92dq0d .tatsu-action-button:hover span{color: rgba(0,0,0,1) ;}</style></div></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div></div><style>.tatsu-row > .tatsu-gso399v44yfjatr.tatsu-column{width: 100%;}.tatsu-gso399v44yfjatr.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: none;}.tatsu-gso399v44yfjatr > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-gso399v44yfjatr > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-gso399v44yfjatr > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-gso399v44yfjatr > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}</style></div></div></div><div class="tatsu-row-wrap tatsu-wrap tatsu-row-one-col tatsu-row-has-one-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-gso4e6n99heqyl3l" ><div class="tatsu-row " ><div class="tatsu-column tatsu-column-no-bg tatsu-one-col tatsu-column-image-none tatsu-column-effect-none tatsu-gso4e6n9d0dgxnue" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class="tatsu-module tatsu-text-block-wrap tatsu-gso4e6n9fa6zs29r "><div class="tatsu-text-inner tatsu-align-center clearfix" ><style>.tatsu-gso4e6n9fa6zs29r.tatsu-text-block-wrap .tatsu-text-inner{width: 100%;text-align: left;}</style><p><span style="color: #000000;">.</span></p><p style="text-align: center;"><span style="font-size: 36pt; color: #000000;">How It Works</span></p></div></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div></div><style>.tatsu-row > .tatsu-gso4e6n9d0dgxnue.tatsu-column{width: 100%;}.tatsu-gso4e6n9d0dgxnue.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: none;}.tatsu-gso4e6n9d0dgxnue > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-gso4e6n9d0dgxnue > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-gso4e6n9d0dgxnue > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-gso4e6n9d0dgxnue > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}</style></div></div></div><div class="tatsu-row-wrap tatsu-wrap tatsu-row-has-three-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-gso4e6n9hk7vnbl4" ><div class="tatsu-row " ><div class="tatsu-column tatsu-column-no-bg tatsu-one-third tatsu-column-image-none tatsu-column-effect-none tatsu-gso4e6n9kvaq8lzq" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class="tatsu-single-image tatsu-module tatsu-external-image tatsu-gso4e6natb9zcrr4 " ><div class="tatsu-single-image-inner " style="" ><div class = "tatsu-single-image-padding-wrap" style = "" ></div><img class = "tatsu-gradient-border" src = "https://odsc.com/wp-content/uploads/2021/03/auditorium_eventx.jpg" alt =" " /></div><style>.tatsu-gso4e6natb9zcrr4{margin: 0px 0px 0px 0px;}.tatsu-gso4e6natb9zcrr4 .tatsu-single-image-inner{border-style: solid;max-width: 100%;}.tatsu-gso4e6natb9zcrr4.tatsu-single-image{transform: translate3d(0px,0px, 0);}</style></div><div class="tatsu-empty-space tatsu-gso4e6nau270f9rf " ><style>.tatsu-gso4e6nau270f9rf.tatsu-empty-space{height: 30px;}</style></div><div class="tatsu-single-image tatsu-module tatsu-external-image tatsu-gso4e6nbfy9rw3p0 " ><div class="tatsu-single-image-inner " style="" ><div class = "tatsu-single-image-padding-wrap" style = "" ></div><img class = "tatsu-gradient-border" src = "https://odsc.com/wp-content/uploads/2021/03/lounge_eventx.jpg" alt =" " /></div><style>.tatsu-gso4e6nbfy9rw3p0{margin: 0px 0px 0px 0px;}.tatsu-gso4e6nbfy9rw3p0 .tatsu-single-image-inner{border-style: solid;max-width: 100%;}.tatsu-gso4e6nbfy9rw3p0.tatsu-single-image{transform: translate3d(0px,0px, 0);}</style></div><div class="tatsu-empty-space tatsu-gso4e6nbgd31g2od " ><style>.tatsu-gso4e6nbgd31g2od.tatsu-empty-space{height: 30px;}</style></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div></div><style>.tatsu-row > .tatsu-gso4e6n9kvaq8lzq.tatsu-column{width: 33.33%;}.tatsu-gso4e6n9kvaq8lzq.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: none;}.tatsu-gso4e6n9kvaq8lzq > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-gso4e6n9kvaq8lzq > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-gso4e6n9kvaq8lzq > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-gso4e6n9kvaq8lzq > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-gso4e6n9kvaq8lzq.tatsu-column{width: 100%;}}</style></div><div class="tatsu-column tatsu-column-no-bg tatsu-one-third tatsu-column-image-none tatsu-column-effect-none tatsu-gso4e6nbi87etd5k" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><ul class="tatsu-module tatsu-list tatsu-gso4e6nbmk2nyjzc tatsu-list-vertical-align-center tatsu-lists-icon" ><style>.tatsu-gso4e6nbmk2nyjzc .tatsu-list-content{margin: 0 0 12px 0px;}.tatsu-gso4e6nbmk2nyjzc .tatsu-list-content::before, .tatsu-gso4e6nbmk2nyjzc .tatsu-icon{color: rgba(34,147,215,1) ;}</style><li class="tatsu-list-content tatsu-gso4e6nbovelqyqm " ><div class="tatsu-list-icon-wrap" ><i class="tatsu-icon icon-icon_check "></i></div><div class="tatsu-list-inner"><p><span style="color: #000000;">Virtual conference experience includes networking lounge area, speaker auditorium, expo halls, and prizes</span></p></div><style>.tatsu-gso4e6nbovelqyqm .tatsu-icon, .tatsu-gso4e6nbovelqyqm.tatsu-list-content::before{color: #00bcdd ;}.tatsu-gso4e6nbovelqyqm{border-style: solid;}</style></li><li class="tatsu-list-content tatsu-gso4e6nbqtdjzeqg " ><div class="tatsu-list-icon-wrap" ><i class="tatsu-icon icon-icon_check "></i></div><div class="tatsu-list-inner"><p><span style="color: #000000;">Access multiple livestream tracks on Tuesday, Wednesday, Thursday</span></p></div><style>.tatsu-gso4e6nbqtdjzeqg .tatsu-icon, .tatsu-gso4e6nbqtdjzeqg.tatsu-list-content::before{color: #00bcdd ;}.tatsu-gso4e6nbqtdjzeqg{border-style: solid;}</style></li><li class="tatsu-list-content tatsu-gso4e6nbrv9m6g5j " ><div class="tatsu-list-icon-wrap" ><i class="tatsu-icon icon-icon_check "></i></div><div class="tatsu-list-inner"><p><span style="color: #000000;">Switch between sessions or tracks as your interests dictate</span></p></div><style>.tatsu-gso4e6nbrv9m6g5j .tatsu-icon, .tatsu-gso4e6nbrv9m6g5j.tatsu-list-content::before{color: #00bcdd ;}.tatsu-gso4e6nbrv9m6g5j{border-style: solid;}</style></li><li class="tatsu-list-content tatsu-gso4e6nbssbkfwj7 " ><div class="tatsu-list-icon-wrap" ><i class="tatsu-icon icon-icon_check "></i></div><div class="tatsu-list-inner"><p><span style="color: #000000;">Multiple focus areas including deep learning, machine learning, NLP, research frontiers, AI X for business, and more</span></p></div><style>.tatsu-gso4e6nbssbkfwj7 .tatsu-icon, .tatsu-gso4e6nbssbkfwj7.tatsu-list-content::before{color: #00bcdd ;}.tatsu-gso4e6nbssbkfwj7{border-style: solid;}</style></li><li class="tatsu-list-content tatsu-gso4e6nbtr5kb6rm " ><div class="tatsu-list-icon-wrap" ><i class="tatsu-icon icon-icon_check "></i></div><div class="tatsu-list-inner"><p><span style="color: #000000;">Sessions you missed can be viewed on demand at your leisure</span></p></div><style>.tatsu-gso4e6nbtr5kb6rm .tatsu-icon, .tatsu-gso4e6nbtr5kb6rm.tatsu-list-content::before{color: #00bcdd ;}.tatsu-gso4e6nbtr5kb6rm{border-style: solid;}</style></li></ul></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div></div><style>.tatsu-row > .tatsu-gso4e6nbi87etd5k.tatsu-column{width: 33.33%;}.tatsu-gso4e6nbi87etd5k.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: none;}.tatsu-gso4e6nbi87etd5k > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-gso4e6nbi87etd5k > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-gso4e6nbi87etd5k > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-gso4e6nbi87etd5k > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-gso4e6nbi87etd5k.tatsu-column{width: 100%;}}</style></div><div class="tatsu-column tatsu-column-no-bg tatsu-one-third tatsu-column-image-none tatsu-column-effect-none tatsu-gso4e6nbwafc2z3a" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><ul class="tatsu-module tatsu-list tatsu-gso4e6nc0d8cgo1p tatsu-list-vertical-align-center tatsu-lists-icon" ><style>.tatsu-gso4e6nc0d8cgo1p .tatsu-list-content{margin: 0 0 12px 0px;}.tatsu-gso4e6nc0d8cgo1p .tatsu-list-content::before, .tatsu-gso4e6nc0d8cgo1p .tatsu-icon{color: rgba(34,147,215,1) ;}</style><li class="tatsu-list-content tatsu-gso4e6nc3m7ux10j " ><div class="tatsu-list-icon-wrap" ><i class="tatsu-icon icon-icon_check "></i></div><div class="tatsu-list-inner"><p><span style="color: #000000;">Engage virtually with fellow attendees, speakers, and Expo partners</span></p></div><style>.tatsu-gso4e6nc3m7ux10j .tatsu-icon, .tatsu-gso4e6nc3m7ux10j.tatsu-list-content::before{color: #00bcdd ;}.tatsu-gso4e6nc3m7ux10j{border-style: solid;}</style></li><li class="tatsu-list-content tatsu-gso4e6nc4u43shp7 " ><div class="tatsu-list-icon-wrap" ><i class="tatsu-icon icon-icon_check "></i></div><div class="tatsu-list-inner"><p><span style="color: #000000;">Participate in Q&A sessions with your speaker over live chat</span></p></div><style>.tatsu-gso4e6nc4u43shp7 .tatsu-icon, .tatsu-gso4e6nc4u43shp7.tatsu-list-content::before{color: #00bcdd ;}.tatsu-gso4e6nc4u43shp7{border-style: solid;}</style></li><li class="tatsu-list-content tatsu-gso4e6nc5tfnjjrj " ><div class="tatsu-list-icon-wrap" ><i class="tatsu-icon icon-icon_check "></i></div><div class="tatsu-list-inner"><p><span style="color: #000000;">Directly download slides and other session materials</span></p></div><style>.tatsu-gso4e6nc5tfnjjrj .tatsu-icon, .tatsu-gso4e6nc5tfnjjrj.tatsu-list-content::before{color: #00bcdd ;}.tatsu-gso4e6nc5tfnjjrj{border-style: solid;}</style></li><li class="tatsu-list-content tatsu-gso4e6nc6qbxbg6j " ><div class="tatsu-list-icon-wrap" ><i class="tatsu-icon icon-icon_check "></i></div><div class="tatsu-list-inner"><p><span style="color: #000000;">(Training only) Access training and workshops prerequisites, notebooks, and other materials prior to training session starting</span></p></div><style>.tatsu-gso4e6nc6qbxbg6j .tatsu-icon, .tatsu-gso4e6nc6qbxbg6j.tatsu-list-content::before{color: #00bcdd ;}.tatsu-gso4e6nc6qbxbg6j{border-style: solid;}</style></li><li class="tatsu-list-content tatsu-gso4e6nc7nfor0ff " ><div class="tatsu-list-icon-wrap" ><i class="tatsu-icon icon-icon_check "></i></div><div class="tatsu-list-inner"><p><span style="color: #000000;">(Training only) Access hands-on training and workshops with instructor-led code labs and notebooks.</span></p></div><style>.tatsu-gso4e6nc7nfor0ff .tatsu-icon, .tatsu-gso4e6nc7nfor0ff.tatsu-list-content::before{color: #00bcdd ;}.tatsu-gso4e6nc7nfor0ff{border-style: solid;}</style></li></ul></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div></div><style>.tatsu-row > .tatsu-gso4e6nbwafc2z3a.tatsu-column{width: 33.33%;}.tatsu-gso4e6nbwafc2z3a.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: none;}.tatsu-gso4e6nbwafc2z3a > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-gso4e6nbwafc2z3a > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-gso4e6nbwafc2z3a > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-gso4e6nbwafc2z3a > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-gso4e6nbwafc2z3a.tatsu-column{width: 100%;}}</style></div></div></div><div class="tatsu-row-wrap tatsu-wrap tatsu-row-one-col tatsu-row-has-one-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-SJgFgga4RO" ><div class="tatsu-row " ><div class="tatsu-column tatsu-bg-overlay tatsu-one-col tatsu-column-image-none tatsu-column-effect-none tatsu-S1Yxl64Cu" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class="tatsu-module tatsu-call-to-action tatsu-clearfix tatsu-gso4euj49j9qp7a2 " ><div class="tatsu-cta-inner"><h4 class="tatsu-action-content" >Interested? Don't doubt - explore ODSC West 2022 Conference</h4><a class="mediumbtn tatsu-button rounded tatsu-action-button " href="https://odsc.com/california/#register" data-gdpr-atts={} ><span>Register and save 60%</span></a><style>.tatsu-gso4euj49j9qp7a2.tatsu-call-to-action{background-color: rgba(0,0,0,1);}.tatsu-gso4euj49j9qp7a2 .tatsu-action-content{color: #ffffff ;}.tatsu-gso4euj49j9qp7a2 .tatsu-action-button{background: rgba(255,0,0,1);border-color: rgba(131,131,131,1); }.tatsu-gso4euj49j9qp7a2 .tatsu-action-button:hover{background: #ffffff;border-color: rgba(0,0,0,1); }.tatsu-gso4euj49j9qp7a2 .tatsu-action-button span{color: rgba(236,224,224,1) ;}.tatsu-gso4euj49j9qp7a2 .tatsu-action-button:hover span{color: #323840 ;}</style></div></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div><div class="tatsu-overlay tatsu-column-overlay tatsu-animate-none" ></div></div><style>.tatsu-row > .tatsu-S1Yxl64Cu.tatsu-column{width: 100%;}.tatsu-S1Yxl64Cu.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: normal;}.tatsu-S1Yxl64Cu > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-S1Yxl64Cu > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-S1Yxl64Cu > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-S1Yxl64Cu > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width:1377px) {.tatsu-row > .tatsu-S1Yxl64Cu.tatsu-column{width: 100%;}}@media only screen and (min-width:768px) and (max-width: 1024px) {.tatsu-row > .tatsu-S1Yxl64Cu.tatsu-column{width: 100%;}}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-S1Yxl64Cu.tatsu-column{width: 100%;}}</style></div></div></div></div><div class="tatsu-section-background-wrap"><div class = "tatsu-section-background" ></div></div><style>.tatsu-gso4e6n9418nmat9 .tatsu-section-background{background-image: url(https://odsc.com/wp-content/uploads/2019/02/распас.png);background-repeat: no-repeat;background-attachment: scroll;background-position: top left;background-size: cover;}.tatsu-gso4e6n9418nmat9 .tatsu-bg-blur{background-repeat: no-repeat;background-attachment: scroll;background-position: top left;background-size: cover;}.tatsu-gso4e6n9418nmat9 .tatsu-section-pad{padding: 10px 0px 30px 0px ;}.tatsu-gso4e6n9418nmat9 > .tatsu-bottom-divider{z-index: 9999;}.tatsu-gso4e6n9418nmat9 > .tatsu-top-divider{z-index: 9999;}</style></div><div id="newsletter" class="tatsu-gyzikjh3qv36o8mu tatsu-section tatsu-clearfix" data-title="" data-headerscheme="background--dark"><div class='tatsu-section-pad clearfix' data-padding='{"d":"30px 0% 30px 0%"}' data-padding-top='30px'><div class="tatsu-row-wrap tatsu-wrap tatsu-row-has-one-half tatsu-row-has-two-cols tatsu-medium-gutter tatsu-reg-cols tatsu-clearfix tatsu-gyzikjh3w5anw00r" ><div class="tatsu-row " ><div class="tatsu-column tatsu-column-no-bg tatsu-one-half tatsu-column-image-none tatsu-column-effect-none tatsu-gyzikjh416doys0y" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class="special-heading-wrap style1 oshine-module tatsu-gyzikjh44lnqswn " ><div class="special-heading align-center"><h3 class="special-h-tag" >ODSC Newsletter</h3><div class="sub-title margin-bottom "><p><span style="color: #ffffff;">Stay current with the latest news and updates in open source data science. In addition, we’ll inform you about our many upcoming Virtual and in person events in Boston, NYC, Sao Paulo, San Francisco, and London. And keep a lookout for special discount codes, only available to our newsletter subscribers!</span></p></div></div><style>.tatsu-gyzikjh44lnqswn .special-h-tag{color: #ffffff ;}</style></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div></div><style>.tatsu-row > .tatsu-gyzikjh416doys0y.tatsu-column{width: 50%;}.tatsu-gyzikjh416doys0y.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: none;}.tatsu-gyzikjh416doys0y > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-gyzikjh416doys0y > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-gyzikjh416doys0y > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-gyzikjh416doys0y > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-gyzikjh416doys0y.tatsu-column{width: 100%;}}</style></div><div class="tatsu-column tatsu-column-no-bg tatsu-one-half tatsu-column-image-none tatsu-column-effect-none tatsu-gyzikjh48xg65mw" data-parallax-speed="0" style=""><div class="tatsu-column-inner " ><div class="tatsu-column-pad-wrap"><div class="tatsu-column-pad" ><div class="tatsu-module tatsu-text-block-wrap tatsu-gyzikjh4d11xyfds "><div class="tatsu-text-inner tatsu-align-center clearfix" ><style>.tatsu-gyzikjh4d11xyfds.tatsu-text-block-wrap .tatsu-text-inner{text-align: left;}</style><div id="form-wrapper" style="max-width: 500px; margin: auto;"> <script charset="utf-8" type="text/javascript" src="//js.hsforms.net/forms/v2-legacy.js"></script> <script charset="utf-8" type="text/javascript" src="//js.hsforms.net/forms/v2.js"></script> <script> hbspt.forms.create({ portalId: "1865444", formId: "e70a0152-a3d8-4863-b9be-c1b4e591cbfc" }); </script></div><br /></div></div></div></div><div class = "tatsu-column-bg-image-wrap"><div class = "tatsu-column-bg-image" ></div></div></div><style>.tatsu-row > .tatsu-gyzikjh48xg65mw.tatsu-column{width: 50%;}.tatsu-gyzikjh48xg65mw.tatsu-column > .tatsu-column-inner > .tatsu-column-overlay{mix-blend-mode: none;}.tatsu-gyzikjh48xg65mw > .tatsu-column-inner > .tatsu-top-divider{z-index: 9999;}.tatsu-gyzikjh48xg65mw > .tatsu-column-inner > .tatsu-bottom-divider{z-index: 9999;}.tatsu-gyzikjh48xg65mw > .tatsu-column-inner > .tatsu-left-divider{z-index: 9999;}.tatsu-gyzikjh48xg65mw > .tatsu-column-inner > .tatsu-right-divider{z-index: 9999;}@media only screen and (max-width: 767px) {.tatsu-row > .tatsu-gyzikjh48xg65mw.tatsu-column{width: 100%;}}</style></div></div></div></div><div class="tatsu-section-background-wrap"><div class = "tatsu-section-background" ></div></div><style>.tatsu-gyzikjh3qv36o8mu.tatsu-section{background-color: rgba(0,0,0,1);}.tatsu-gyzikjh3qv36o8mu{border-width: 0px 0px px 0px;border-style: solid;}.tatsu-gyzikjh3qv36o8mu .tatsu-section-pad{padding: 30px 0% 30px 0%;}.tatsu-gyzikjh3qv36o8mu > .tatsu-bottom-divider{z-index: 9999;}.tatsu-gyzikjh3qv36o8mu > .tatsu-top-divider{z-index: 9999;}</style></div></div> <!-- End Page Content --></section></div></div><footer id="bottom-widgets"><div id="bottom-widgets-wrap" class="be-wrap be-row clearfix"><div class="one-fourth column-block clearfix"><div class="widget_text widget"><h6>Open Data Science</h6><div class="textwidget"><div class="textwidget"><div class="tatsu-module tatsu-normal-icon tatsu-icon-shortcode align-none tatsu-goka2it139bzrcg7 "> <style>.tatsu-goka2it139bzrcg7 .tatsu-icon{background-color: #00aced;color: #ffffff ;border-color: #ffffff; }.tatsu-goka2it139bzrcg7 .tatsu-icon:hover{background-color: #00aced;color: #ffffff ;}</style><p> </p></div><div class="tatsu-module tatsu-normal-icon tatsu-icon-shortcode align-none tatsu-goka2it16zd3n28v "> <style>.tatsu-goka2it16zd3n28v .tatsu-icon{background-color: #3B5998;color: #ffffff ;border-color: #ffffff; }.tatsu-goka2it16zd3n28v .tatsu-icon:hover{background-color: #3B5998;color: #ffffff ;}</style><p> </p></div><div class="tatsu-module tatsu-normal-icon tatsu-icon-shortcode align-none tatsu-goka2it19262p6uh "> <style>.tatsu-goka2it19262p6uh .tatsu-icon{background-color: #bb0000;color: #ffffff ;border-color: #ffffff; }.tatsu-goka2it19262p6uh .tatsu-icon:hover{background-color: #bb0000;color: #ffffff ;}</style><p> </p></div><p>Open Data Science<br /> One Broadway<br /> Cambridge, MA 02142<br /> info@odsc.com</p></div></div></div></div><div class="one-fourth column-block clearfix"><div class="widget_nav_menu widget"><h6>Menu</h6><div class="menu-footer_nav-container"><ul id="menu-footer_nav" class="menu"><li id="menu-item-34705" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-34705"><a href="/boston/partner-odsc-east/" title="Partner with ODSC">Partner with ODSC</a></li><li id="menu-item-1656" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-1656"><a href="https://www.opendatascience.com/" title="Blog">Blog</a></li><li id="menu-item-1657" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-1657"><a href="https://aiplus.training/" title="Training">Training</a></li><li id="menu-item-1655" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-1655"><a href="https://jobs.opendatascience.com/" title="Jobs">Jobs</a></li><li id="menu-item-14010" class="menu-item menu-item-type-post_type menu-item-object-page menu-item-14010"><a href="https://odsc.com/faq/" title="FAQ">FAQ</a></li></ul></div></div></div><div class="one-fourth column-block clearfix"><div class="widget_nav_menu widget"><h6>Conferences</h6><div class="menu-conf_footer-container"><ul id="menu-conf_footer" class="menu"><li id="menu-item-27912" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-27912"><a href="/boston/" title="East 2025">East 2025</a></li><li id="menu-item-45631" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-45631"><a href="/europe/" title="Europe 2025">Europe 2025</a></li><li id="menu-item-19688" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-19688"><a href="/apac/" title="APAC 2025">APAC 2025</a></li><li id="menu-item-1661" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-1661"><a href="/california/" title="West 2025">West 2025</a></li></ul></div></div></div><div class="one-fourth column-block clearfix"><div class="widget_nav_menu widget"><h6>Extras</h6><div class="menu-extras-container"><ul id="menu-extras" class="menu"><li id="menu-item-1663" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-1663"><a href="https://opendatascience.com/newsletter/?__hstc=19222759.9747cccda111769b206e3552831bc327.1533925250574.1535328146852.1535383810113.51&__hssc=19222759.10.1535383810113&__hsfp=1117387146" title="Newsletter">Newsletter</a></li><li id="menu-item-58125" class="menu-item menu-item-type-custom menu-item-object-custom menu-item-58125"><a href="/about-open-data-science-conference/" title="About">About</a></li><li id="menu-item-1666" class="menu-item menu-item-type-post_type menu-item-object-page menu-item-1666"><a href="https://odsc.com/code-of-conduct/" title="Code of Conduct">Code of Conduct</a></li><li id="menu-item-1667" class="menu-item menu-item-type-post_type menu-item-object-page menu-item-1667"><a href="https://odsc.com/privacy-policy/" title="Privacy Policy">Privacy Policy</a></li></ul></div></div></div></div></footer><footer id="footer" class="layout-wide"> <span class="footer-border "></span><div id="footer-wrap" class=" style1 clearfix"><div class="footer-left-area"></div><div class="footer-center-area"><div class="footer-content-inner-center"> Copyright ODSC 2024. All Rights Reserved</div></div><div class="footer-right-area"></div></div></footer></div><div class="loader page-loader"><div class="loader-style1-wrap"><div class="loader-style1-double-bounce1"></div><div class="loader-style1-double-bounce2"></div></div></div> <a href="#" id="back-to-top" class="layout-wide"><i class="font-icon icon-arrow_carrot-up"></i></a></div> <input type="hidden" id="ajax_url" value="https://odsc.com/wp-admin/admin-ajax.php" /><div id="gdpr-popup" class="be-gdpr-popup mfp-hide " data-rel="gdpr-popup-gallery" ><div class="be-gdpr-modal" ><div class="be-modal-content-wrapper" ><div class="be-gdpr-modal-heading" >Privacy Settings</div><div class="be-gdpr-modal-desc" > We use cookies to enhance your experience while using our website. If you are using our Services via a browser you can restrict, block or remove cookies through your web browser settings. We also use content and scripts from third parties that may use tracking technologies. You can selectively provide your consent below to allow such third party embeds. For complete information about the cookies we use, data we collect and how we process them, please check our <a target="_blank" href="">Privacy Policy</a></div><div class="be-gdpr-modal-items" ><div class="be-gdpr-modal-item" ><div class="be-gdpr-modal-item-head" >Youtube</div><div class="be-gdpr-modal-item-desc" >Consent to display content from - Youtube</div><div class="be-gdpr-modal-item-switch"> <label class="switch be-modal-switch"> <input class="be-gdpr-switch-input" value="youtube" type="checkbox"> <span class="slider round"></span> </label></div></div><div class="be-gdpr-modal-item" ><div class="be-gdpr-modal-item-head" >Vimeo</div><div class="be-gdpr-modal-item-desc" >Consent to display content from - Vimeo</div><div class="be-gdpr-modal-item-switch"> <label class="switch be-modal-switch"> <input class="be-gdpr-switch-input" value="vimeo" type="checkbox"> <span class="slider round"></span> </label></div></div><div class="be-gdpr-modal-item" ><div class="be-gdpr-modal-item-head" >Google Maps</div><div class="be-gdpr-modal-item-desc" >Consent to display content from - Google</div><div class="be-gdpr-modal-item-switch"> <label class="switch be-modal-switch"> <input class="be-gdpr-switch-input" value="gmaps" type="checkbox"> <span class="slider round"></span> </label></div></div></div></div><div class="be-gdpr-modal-footer" ><div class="be-gdpr-modal-save-btn" onClick="gdprSaveBtnClick(event);" >Save</div></div></div></div> <script type="text/javascript"> adroll_adv_id = "MN7XH5FMX5AODAUTQ7LEYH"; adroll_pix_id = "4WF3QPQPGRHLJNYPUIJSDC"; (function () { var _onload = function(){ if (document.readyState && !/loaded|complete/.test(document.readyState)){setTimeout(_onload, 10);return} if (!window.__adroll_loaded){__adroll_loaded=true;setTimeout(_onload, 50);return} var scr = document.createElement("script"); var host = (("https:" == document.location.protocol) ? "https://s.adroll.com" : "http://a.adroll.com"); scr.setAttribute('async', 'true'); scr.type = "text/javascript"; scr.src = host + "/j/roundtrip.js"; ((document.getElementsByTagName('head') || [null])[0] || document.getElementsByTagName('script')[0].parentNode).appendChild(scr); }; if (window.addEventListener) {window.addEventListener('load', _onload, false);} else {window.attachEvent('onload', _onload)} }()); </script> <script type="text/javascript"> _linkedin_partner_id = "44953"; window._linkedin_data_partner_ids = window._linkedin_data_partner_ids || []; window._linkedin_data_partner_ids.push(_linkedin_partner_id); </script><script type="text/javascript"> (function(l) { if (!l){window.lintrk = function(a,b){window.lintrk.q.push([a,b])}; window.lintrk.q=[]} var s = document.getElementsByTagName("script")[0]; var b = document.createElement("script"); b.type = "text/javascript";b.async = true; b.src = "https://snap.licdn.com/li.lms-analytics/insight.min.js"; s.parentNode.insertBefore(b, s);})(window.lintrk); </script> <noscript> <img height="1" width="1" style="display:none;" alt="" src="https://px.ads.linkedin.com/collect/?pid=44953&fmt=gif" /> </noscript><script type="text/javascript"> adroll_adv_id = "MN7XH5FMX5AODAUTQ7LEYH"; adroll_pix_id = "4WF3QPQPGRHLJNYPUIJSDC"; (function () { var _onload = function(){ if (document.readyState && !/loaded|complete/.test(document.readyState)){setTimeout(_onload, 10);return} if (!window.__adroll_loaded){__adroll_loaded=true;setTimeout(_onload, 50);return} var scr = document.createElement("script"); var host = (("https:" == document.location.protocol) ? "https://s.adroll.com" : "http://a.adroll.com"); scr.setAttribute('async', 'true'); scr.type = "text/javascript"; scr.src = host + "/j/roundtrip.js"; ((document.getElementsByTagName('head') || [null])[0] || document.getElementsByTagName('script')[0].parentNode).appendChild(scr); }; if (window.addEventListener) {window.addEventListener('load', _onload, false);} else {window.attachEvent('onload', _onload)} }()); </script> <script type='text/javascript'> window.__lo_site_id = 303225; (function() { var wa = document.createElement('script'); wa.type = 'text/javascript'; wa.async = true; wa.src = 'https://d10lpsik1i8c69.cloudfront.net/w.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(wa, s); })(); </script><div id="gallery" class="pswp" tabindex="-1" role="dialog" aria-hidden="true"><div class="pswp__bg"></div><div class="pswp__scroll-wrap"><div class="pswp__container"><div class="pswp__item"></div><div class="pswp__item"></div><div class="pswp__item"></div></div><div class="pswp__ui pswp__ui--hidden"><div class="pswp__top-bar"><div class="pswp__counter"></div> <button class="pswp__button pswp__button--close" title="Close (Esc)"></button> <button class="pswp__button pswp__button--share" title="Share"></button> <button class="pswp__button pswp__button--fs" title="Toggle fullscreen"></button> <button class="pswp__button pswp__button--zoom" title="Zoom in/out"></button><div class="pswp__preloader"><div class="pswp__preloader__icn"><div class="pswp__preloader__cut"><div class="pswp__preloader__donut"></div></div></div></div></div> <!--<div class="pswp__loading-indicator"><div class="pswp__loading-indicator__line"></div></div> --><div class="pswp__share-modal pswp__share-modal--hidden pswp__single-tap"><div class="pswp__share-tooltip"> <!-- <a href="#" class="pswp__share--facebook"></a> <a href="#" class="pswp__share--twitter"></a> <a href="#" class="pswp__share--pinterest"></a> <a href="#" download class="pswp__share--download"></a> --></div></div> <button class="pswp__button pswp__button--arrow--left" title="Previous (arrow left)"></button> <button class="pswp__button pswp__button--arrow--right" title="Next (arrow right)"></button><div class="pswp__caption"><div class="pswp__caption__center"></div></div></div></div></div><div id="pum-60769" role="dialog" aria-modal="false" class="pum pum-overlay pum-theme-47575 pum-theme-lightbox popmake-overlay click_open" data-popmake="{"id":60769,"slug":"tree","theme_id":47575,"cookies":[],"triggers":[{"type":"click_open","settings":{"cookie_name":"","extra_selectors":".path1 a"}}],"mobile_disabled":true,"tablet_disabled":null,"meta":{"display":{"stackable":false,"overlay_disabled":false,"scrollable_content":false,"disable_reposition":false,"size":"normal","responsive_min_width":"0%","responsive_min_width_unit":false,"responsive_max_width":"100%","responsive_max_width_unit":false,"custom_width":"640px","custom_width_unit":false,"custom_height":"380px","custom_height_unit":false,"custom_height_auto":false,"location":"center top","position_from_trigger":false,"position_top":"100","position_left":"0","position_bottom":"0","position_right":"0","position_fixed":false,"animation_type":"fade","animation_speed":"350","animation_origin":"center top","overlay_zindex":false,"zindex":"1999999999"},"close":{"text":"","button_delay":"0","overlay_click":false,"esc_press":false,"f4_press":false},"click_open":[]}}"><div id="popmake-60769" class="pum-container popmake theme-47575 pum-responsive pum-responsive-normal responsive size-normal"><div class="pum-content popmake-content" tabindex="0"><div id="tree-container"></div> <style> .pum-theme-47575 .pum-container, .pum-theme-lightbox .pum-container {<br /> padding: 50px;<br /> border-radius: 3px;<br /> border: 8px solid #262424;<br /> box-shadow: 0px 0px 30px 0px rgb(2 2 2);<br /> background-color: rgb(0 0 0);<br /> }</p> </style></div> <button type="button" class="pum-close popmake-close" aria-label="Close"> × </button></div></div> <!-- GTM Container placement set to footer --> <!-- Google Tag Manager (noscript) --> <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-M62L34N" height="0" width="0" style="display:none;visibility:hidden" aria-hidden="true"></iframe></noscript> <!-- End Google Tag Manager (noscript) --> <script> jQuery(document).ready(function($){ var links = $('#eventin-external-event-list').find('a'); $(links).attr('target', '_blank'); }); </script> <style id='etn-custom-css-inline-css' type='text/css'> .etn-event-single-content-wrap .etn-event-meta .etn-event-category span, .etn-event-item .etn-event-footer .etn-atend-btn .etn-btn-border, .etn-btn.etn-btn-border, .attr-btn-primary.etn-btn-border, .etn-attendee-form .etn-btn.etn-btn-border, .etn-ticket-widget .etn-btn.etn-btn-border, .etn-settings-dashboard .button-primary.etn-btn-border, .etn-single-speaker-item .etn-speaker-content a:hover, .etn-event-style2 .etn-event-date, .etn-event-style3 .etn-event-content .etn-title a:hover, .event-tab-wrapper ul li a.etn-tab-a, .etn-speaker-item.style-3:hover .etn-speaker-content .etn-title a, .etn-variable-ticket-widget .ticket-header, .events_calendar_list .calendar-event-details:hover .calendar-event-title, .etn-event-item:hover .etn-title a, .etn-recurring-widget .etn-date-text, .etn-event-header ul li i { color: #00bcdd; } .etn-event-item .etn-event-category span, .etn-btn, .attr-btn-primary, .etn-attendee-form .etn-btn, .etn-ticket-widget .etn-btn, .schedule-list-1 .schedule-header, .speaker-style4 .etn-speaker-content .etn-title a, .etn-speaker-details3 .speaker-title-info, .etn-event-slider .swiper-pagination-bullet, .etn-speaker-slider .swiper-pagination-bullet, .etn-event-slider .swiper-button-next, .etn-event-slider .swiper-button-prev, .etn-speaker-slider .swiper-button-next, .etn-speaker-slider .swiper-button-prev, .etn-single-speaker-item .etn-speaker-thumb .etn-speakers-social a, .etn-event-header .etn-event-countdown-wrap .etn-count-item, .schedule-tab-1 .etn-nav li a.etn-active, .schedule-list-wrapper .schedule-listing.multi-schedule-list .schedule-slot-time, .etn-speaker-item.style-3 .etn-speaker-content .etn-speakers-social a, .event-tab-wrapper ul li a.etn-tab-a.etn-active, .etn-btn, button.etn-btn.etn-btn-primary, .etn-schedule-style-3 ul li:before, .etn-zoom-btn, .cat-radio-btn-list [type=radio]:checked+label:after, .cat-radio-btn-list [type=radio]:not(:checked)+label:after, .etn-default-calendar-style .fc-button:hover, .etn-default-calendar-style .fc-state-highlight, .etn-calender-list a:hover, .events_calendar_standard .cat-dropdown-list select, .etn-event-banner-wrap, .events_calendar_list .calendar-event-details .calendar-event-content .calendar-event-category-wrap .etn-event-category, .etn-variable-ticket-widget .etn-add-to-cart-block, .etn-recurring-event-wrapper #seeMore, .more-event-tag, .etn-settings-dashboard .button-primary{ background-color: #00bcdd; } .etn-event-item .etn-event-footer .etn-atend-btn .etn-btn-border, .etn-btn.etn-btn-border, .attr-btn-primary.etn-btn-border, .etn-attendee-form .etn-btn.etn-btn-border, .etn-ticket-widget .etn-btn.etn-btn-border, .event-tab-wrapper ul li a.etn-tab-a, .event-tab-wrapper ul li a.etn-tab-a.etn-active, .etn-schedule-style-3 ul li:after, .etn-default-calendar-style .fc-ltr .fc-basic-view .fc-day-top.fc-today .fc-day-number, .etn-default-calendar-style .fc-button:hover, .etn-variable-ticket-widget .etn-variable-total-price, .etn-settings-dashboard .button-primary.etn-btn-border{ border-color: #00bcdd; } .schedule-tab-wrapper .etn-nav li a.etn-active, .etn-speaker-item.style-3 .etn-speaker-content{ border-bottom-color: #00bcdd; } .schedule-tab-wrapper .etn-nav li a:after, .etn-event-list2 .etn-event-content, .schedule-tab-1 .etn-nav li a.etn-active:after{ border-color: #00bcdd transparent transparent transparent; } .etn-default-calendar-style .fc .fc-daygrid-bg-harness:first-of-type:before{ background-color: #00bcdd2A; } .sidebar .etn-default-calendar-style .fc .fc-daygrid-bg-harness:nth-of-type(1)::before, .left-sidebar .etn-default-calendar-style .fc .fc-daygrid-bg-harness:nth-of-type(1)::before, .right-sidebar .etn-default-calendar-style .fc .fc-daygrid-bg-harness:nth-of-type(1)::before, .widget .etn-default-calendar-style .fc .fc-daygrid-bg-harness:nth-of-type(1)::before, .widgets .etn-default-calendar-style .fc .fc-daygrid-bg-harness:nth-of-type(1)::before, .main-sidebar .etn-default-calendar-style .fc .fc-daygrid-bg-harness:nth-of-type(1)::before, #sidebar .etn-default-calendar-style .fc .fc-daygrid-bg-harness:nth-of-type(1)::before{ background-color: #00bcdd; } .etn-event-item .etn-event-location, .etn-event-tag-list a:hover, .etn-schedule-wrap .etn-schedule-info .etn-schedule-time{ color: #0057c1; } .etn-event-tag-list a:hover{ border-color: #0057c1; } .etn-btn:hover, .attr-btn-primary:hover, .etn-attendee-form .etn-btn:hover, .etn-ticket-widget .etn-btn:hover, .speaker-style4 .etn-speaker-content p, .etn-btn, button.etn-btn.etn-btn-primary:hover, .etn-zoom-btn, .events_calendar_list .calendar-event-details .event-calendar-action .etn-btn, .events_calendar_list .calendar-event-details .event-calendar-action .etn-price.event-calendar-details-btn, .etn-speaker-item.style-3 .etn-speaker-content .etn-speakers-social a:hover, .etn-single-speaker-item .etn-speaker-thumb .etn-speakers-social a:hover, .etn-recurring-event-wrapper #seeMore:hover, .etn-recurring-event-wrapper #seeMore:focus, .etn-settings-dashboard .button-primary:hover{ background-color: #0057c1; } </style> <script defer type="text/javascript" src="https://odsc.com/wp-includes/js/imagesloaded.min.js" id="imagesloaded-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-includes/js/masonry.min.js" id="masonry-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-includes/js/jquery/jquery.masonry.min.js" id="jquery-masonry-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-content/plugins/be-gdpr/public/js/magnificpopup.min.js" id="gdprmagnificpopup-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-content/plugins/awsm-team-pro/js/team.min.js" id="awsm-team-js"></script> <script type="text/javascript" id="leadin-script-loader-js-js-extra"> /* <![CDATA[ */ var leadin_wordpress = {"userRole":"visitor","pageType":"page","leadinPluginVersion":"11.1.66"}; /* ]]> */ </script> <script defer type="text/javascript" src="https://js.hs-scripts.com/1865444.js?integration=WordPress" id="leadin-script-loader-js-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-content/plugins/oshine-modules/public/js/vendor/asyncloader.min.js" id="asyncloader-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-includes/js/jquery/ui/accordion.min.js" id="jquery-ui-accordion-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-includes/js/jquery/ui/tabs.min.js" id="jquery-ui-tabs-js"></script> <script type="text/javascript" id="oshine-modules-js-extra"> /* <![CDATA[ */ var oshineModulesConfig = {"pluginUrl":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/","vendorScriptsUrl":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/public\/js\/vendor\/","dependencies":{"asyncloader":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/public\/js\/vendor\/asyncloader.min.js","backgroundcheck":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/public\/js\/vendor\/backgroundcheck.min.js","backgroundposition":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/public\/js\/vendor\/backgroundposition.min.js","be-modules-plugin":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/public\/js\/vendor\/be-modules-plugin.min.js","beslider":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/public\/js\/vendor\/beslider.min.js","countdown":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/public\/js\/vendor\/countdown.min.js","easing":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/public\/js\/vendor\/easing.min.js","easypiechart":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/public\/js\/vendor\/easypiechart.min.js","fitvids":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/public\/js\/vendor\/fitvids.min.js","fullscreenheight":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/public\/js\/vendor\/fullscreenheight.min.js","hoverdir":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/public\/js\/vendor\/hoverdir.min.js","imagesloaded":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/public\/js\/vendor\/imagesloaded.min.js","isotope":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/public\/js\/vendor\/isotope.min.js","justifiedgallery":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/public\/js\/vendor\/justifiedgallery.min.js","magnificpopup":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/public\/js\/vendor\/magnificpopup.min.js","mousewheel":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/public\/js\/vendor\/mousewheel.min.js","owlcarousel":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/public\/js\/vendor\/owlcarousel.min.js","photoswipe":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/public\/js\/vendor\/photoswipe.min.js","resizetoparent":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/public\/js\/vendor\/resizetoparent.min.js","rotate":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/public\/js\/vendor\/rotate.min.js","tilt":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/public\/js\/vendor\/tilt.min.js","typed":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/public\/js\/vendor\/typed.min.js","vivusSVGanimation":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/public\/js\/vendor\/vivusSVGanimation.min.js","waypoints":"https:\/\/odsc.com\/wp-content\/plugins\/oshine-modules\/public\/js\/vendor\/waypoints.min.js"}}; /* ]]> */ </script> <script defer type="text/javascript" src="https://odsc.com/wp-content/plugins/oshine-modules/public/js/oshine-modules.min.js" id="oshine-modules-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-content/plugins/sg-cachepress/assets/js/lazysizes.min.js" id="siteground-optimizer-lazy-sizes-js-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-content/plugins/tatsu/public/js/vendor/es6-promise.auto.min.js" id="es6-promises-polyfill-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-content/plugins/tatsu/public/js/helpers.min.js" id="be-script-helpers-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-content/plugins/tatsu/public/js/vendor/debouncedresize.min.js" id="debouncedresize-js"></script> <script type="text/javascript" id="tatsu-js-extra"> /* <![CDATA[ */ var tatsuFrontendConfig = {"ajax_url":"https:\/\/odsc.com\/wp-admin\/admin-ajax.php","pluginUrl":"https:\/\/odsc.com\/wp-content\/plugins\/tatsu","vendorScriptsUrl":"https:\/\/odsc.com\/wp-content\/plugins\/tatsu\/public\/js\/vendor\/","mapsApiKey":"AIzaSyBApCqI3PUzglTtm3t6AdOqBfEH0sdhN6E","dependencies":{"anime":"https:\/\/odsc.com\/wp-content\/plugins\/tatsu\/public\/js\/vendor\/anime.min.js?ver=3.5.2","asyncloader":"https:\/\/odsc.com\/wp-content\/plugins\/tatsu\/public\/js\/vendor\/asyncloader.min.js?ver=3.5.2","backgroundposition":"https:\/\/odsc.com\/wp-content\/plugins\/tatsu\/public\/js\/vendor\/backgroundposition.min.js?ver=3.5.2","begrid":"https:\/\/odsc.com\/wp-content\/plugins\/tatsu\/public\/js\/vendor\/begrid.min.js?ver=3.5.2","countTo":"https:\/\/odsc.com\/wp-content\/plugins\/tatsu\/public\/js\/vendor\/countTo.min.js?ver=3.5.2","debouncedresize":"https:\/\/odsc.com\/wp-content\/plugins\/tatsu\/public\/js\/vendor\/debouncedresize.min.js?ver=3.5.2","es6-promise.auto":"https:\/\/odsc.com\/wp-content\/plugins\/tatsu\/public\/js\/vendor\/es6-promise.auto.min.js?ver=3.5.2","fitvids":"https:\/\/odsc.com\/wp-content\/plugins\/tatsu\/public\/js\/vendor\/fitvids.min.js?ver=3.5.2","flickity":"https:\/\/odsc.com\/wp-content\/plugins\/tatsu\/public\/js\/vendor\/flickity.min.js?ver=3.5.2","hoverintent":"https:\/\/odsc.com\/wp-content\/plugins\/tatsu\/public\/js\/vendor\/hoverintent.min.js?ver=3.5.2","imagesloaded":"https:\/\/odsc.com\/wp-content\/plugins\/tatsu\/public\/js\/vendor\/imagesloaded.min.js?ver=3.5.2","isotope":"https:\/\/odsc.com\/wp-content\/plugins\/tatsu\/public\/js\/vendor\/isotope.min.js?ver=3.5.2","magnificpopup":"https:\/\/odsc.com\/wp-content\/plugins\/tatsu\/public\/js\/vendor\/magnificpopup.min.js?ver=3.5.2","stickykit":"https:\/\/odsc.com\/wp-content\/plugins\/tatsu\/public\/js\/vendor\/stickykit.min.js?ver=3.5.2","superfish":"https:\/\/odsc.com\/wp-content\/plugins\/tatsu\/public\/js\/vendor\/superfish.min.js?ver=3.5.2","tatsuCarousel":"https:\/\/odsc.com\/wp-content\/plugins\/tatsu\/public\/js\/vendor\/tatsuCarousel.min.js?ver=3.5.2","tatsuColumnParallax":"https:\/\/odsc.com\/wp-content\/plugins\/tatsu\/public\/js\/vendor\/tatsuColumnParallax.min.js?ver=3.5.2","tatsuParallax":"https:\/\/odsc.com\/wp-content\/plugins\/tatsu\/public\/js\/vendor\/tatsuParallax.min.js?ver=3.5.2","tilt":"https:\/\/odsc.com\/wp-content\/plugins\/tatsu\/public\/js\/vendor\/tilt.min.js?ver=3.5.2","typed":"https:\/\/odsc.com\/wp-content\/plugins\/tatsu\/public\/js\/vendor\/typed.min.js?ver=3.5.2","unveil":"https:\/\/odsc.com\/wp-content\/plugins\/tatsu\/public\/js\/vendor\/unveil.min.js?ver=3.5.2","vivus":"https:\/\/odsc.com\/wp-content\/plugins\/tatsu\/public\/js\/vendor\/vivus.min.js?ver=3.5.2"},"slider_icons":{"left":"<svg viewBox=\"0 0 10 16\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\">\r\n<path opacity=\"0.999943\" d=\"M0 0L7 7.1759L0.343184 14\" transform=\"translate(9 15) rotate(-180)\" stroke-width=\"2\"\/>\r\n<\/svg>\r\n","right":"<svg viewBox=\"0 0 10 16\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\">\r\n<path d=\"M0 0L7 7.1759L0.343184 14\" transform=\"translate(1 1)\" stroke-width=\"2\"\/>\r\n<\/svg>\r\n"},"version":"3.5.2","recaptcha_type":"","recaptcha_site_key":"","tatsunonce":"1db3d04d2b"}; /* ]]> */ </script> <script defer type="text/javascript" src="https://odsc.com/wp-content/plugins/tatsu/public/js/tatsu.min.js" id="tatsu-js"></script> <script type="text/javascript" id="audioigniter-js-extra"> /* <![CDATA[ */ var aiStrings = {"play_title":"Play %s","pause_title":"Pause %s","previous":"Previous track","next":"Next track","toggle_list_repeat":"Toggle track listing repeat","toggle_track_repeat":"Toggle track repeat","toggle_list_visible":"Toggle track listing visibility","buy_track":"Buy this track","download_track":"Download this track","volume_up":"Volume Up","volume_down":"Volume Down","open_track_lyrics":"Open track lyrics","set_playback_rate":"Set playback rate","skip_forward":"Skip forward","skip_backward":"Skip backward","shuffle":"Shuffle"}; var aiStats = {"enabled":"","apiUrl":"https:\/\/odsc.com\/wp-json\/audioigniter\/v1"}; /* ]]> */ </script> <script defer type="text/javascript" src="https://odsc.com/wp-content/uploads/siteground-optimizer-assets/audioigniter.min.js" id="audioigniter-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-content/uploads/siteground-optimizer-assets/gtm4wp-form-move-tracker.min.js" id="gtm4wp-form-move-tracker-js"></script> <script type="text/javascript" id="etn-public-js-extra"> /* <![CDATA[ */ var localized_data_obj = {"ajax_url":"https:\/\/odsc.com\/wp-admin\/admin-ajax.php","site_url":"https:\/\/odsc.com","evnetin_pro_active":"1","locale_name":"en-us","start_of_week":"1","expired":"Expired","author_id":"0","scanner_common_msg":"Something went wrong! Please try again.","ticket_scanner_link":"https:\/\/odsc.com\/wp-admin\/edit.php?post_type=etn-attendee","attendee_form_validation_msg":{"email":{"invalid":"Email is not valid","empty":"Please fill the field"},"tel":{"empty":"Please fill the field","invalid":"Invalid phone number","only_number":"Only number allowed"},"text":"Please fill the field","number":"Please input a number","date":"Please fill the field","radio":"Please check the field"},"post_id":"67937"}; /* ]]> */ </script> <script defer type="text/javascript" src="https://odsc.com/wp-content/uploads/siteground-optimizer-assets/etn-public.min.js" id="etn-public-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-content/themes/oshin/oshin/js/vendor/perfect-scrollbar.jquery.min.js" id="custom-scrollbar-js"></script> <script type="text/javascript" id="be-themes-script-js-js-extra"> /* <![CDATA[ */ var oshineThemeConfig = {"vendorScriptsUrl":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/","dependencies":{"asyncloader":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/asyncloader.min.js","backgroundcheck":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/backgroundcheck.min.js","backgroundposition":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/backgroundposition.min.js","classie":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/classie.min.js","easing":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/easing.min.js","fitvids":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/fitvids.min.js","flickity":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/flickity.min.js","fullscreenheight":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/fullscreenheight.min.js","galaxycanvas":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/galaxycanvas.min.js","greensock":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/greensock.min.js","horizontalcarousel":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/horizontalcarousel.min.js","hoverintent":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/hoverintent.min.js","imagesloaded":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/imagesloaded.min.js","isotope":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/isotope.min.js","magnificpopup":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/magnificpopup.min.js","mcustomscrollbar":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/mcustomscrollbar.min.js","modernizr":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/modernizr.min.js","mousewheel":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/mousewheel.min.js","multi_level_menu":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/multi_level_menu.min.js","old_menu_animation":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/old_menu_animation.min.js","page_stack_top":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/page_stack_top.min.js","patterncanvas":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/patterncanvas.min.js","perfect-scrollbar.jquery":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/perfect-scrollbar.jquery.min.js","perspective_navigation":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/perspective_navigation.min.js","request_animation_frame":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/request_animation_frame.min.js","resizetoparent":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/resizetoparent.min.js","scrolltosections":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/scrolltosections.min.js","simplebar":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/simplebar.min.js","simplebar_old":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/simplebar_old.min.js","sticky_sections":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/sticky_sections.min.js","stickykit":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/stickykit.min.js","superfish":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/superfish.min.js","transparentheader":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/transparentheader.min.js","waterdropcanvas":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/waterdropcanvas.min.js","webfont":"https:\/\/odsc.com\/wp-content\/themes\/oshin\/oshin\/js\/vendor\/webfont.min.js"}}; /* ]]> */ </script> <script defer type="text/javascript" src="https://odsc.com/wp-content/themes/oshin/oshin/js/script.min.js" id="be-themes-script-js-js"></script> <script type="text/javascript" id="popup-maker-site-js-extra"> /* <![CDATA[ */ var pum_vars = {"version":"1.20.2","pm_dir_url":"https:\/\/odsc.com\/wp-content\/plugins\/popup-maker\/","ajaxurl":"https:\/\/odsc.com\/wp-admin\/admin-ajax.php","restapi":"https:\/\/odsc.com\/wp-json\/pum\/v1","rest_nonce":null,"default_theme":"47574","debug_mode":"","disable_tracking":"","home_url":"\/","message_position":"top","core_sub_forms_enabled":"1","popups":[],"cookie_domain":"","analytics_route":"analytics","analytics_api":"https:\/\/odsc.com\/wp-json\/pum\/v1"}; var pum_sub_vars = {"ajaxurl":"https:\/\/odsc.com\/wp-admin\/admin-ajax.php","message_position":"top"}; var pum_popups = {"pum-60769":{"triggers":[{"type":"click_open","settings":{"cookie_name":"","extra_selectors":".path1 a"}}],"cookies":[],"disable_on_mobile":true,"disable_on_tablet":false,"atc_promotion":null,"explain":null,"type_section":null,"theme_id":"47575","size":"normal","responsive_min_width":"0%","responsive_max_width":"100%","custom_width":"640px","custom_height_auto":false,"custom_height":"380px","scrollable_content":false,"animation_type":"fade","animation_speed":"350","animation_origin":"center top","open_sound":"none","custom_sound":"","location":"center top","position_top":"100","position_bottom":"0","position_left":"0","position_right":"0","position_from_trigger":false,"position_fixed":false,"overlay_disabled":false,"stackable":false,"disable_reposition":false,"zindex":"1999999999","close_button_delay":"0","fi_promotion":null,"close_on_form_submission":false,"close_on_form_submission_delay":"0","close_on_overlay_click":false,"close_on_esc_press":false,"close_on_f4_press":false,"disable_form_reopen":false,"disable_accessibility":false,"theme_slug":"lightbox","id":60769,"slug":"tree"}}; /* ]]> */ </script> <script defer type="text/javascript" src="//odsc.com/wp-content/uploads/pum/pum-site-scripts.js?defer" id="popup-maker-site-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-content/plugins/popup-maker/assets/js/vendor/mobile-detect.min.js" id="mobile-detect-js"></script> <script type="text/javascript" id="fca_pc_client_js-js-extra"> /* <![CDATA[ */ var fcaPcEvents = []; var fcaPcPost = {"title":"ODSC West Schedule 2024","type":"page","id":"67937","categories":[]}; var fcaPcOptions = {"pixel_types":[],"ajax_url":"https:\/\/odsc.com\/wp-admin\/admin-ajax.php","debug":"","edd_currency":"USD","nonce":"a0e21137e2","utm_support":"","user_parameters":"","edd_enabled":"","edd_delay":"0","woo_enabled":"","woo_delay":"0","video_enabled":""}; /* ]]> */ </script> <script defer type="text/javascript" src="https://odsc.com/wp-content/plugins/facebook-conversion-pixel/pixel-cat.min.js" id="fca_pc_client_js-js"></script> <script defer type="text/javascript" src="https://odsc.com/wp-content/uploads/siteground-optimizer-assets/fca_pc_video_js.min.js" id="fca_pc_video_js-js"></script> <script type="text/javascript" src="https://cdn.onesignal.com/sdks/OneSignalSDK.js" id="remote_sdk-js" async="async" data-wp-strategy="async"></script> <script type="text/javascript" id="gt_widget_script_71620520-js-before"> /* <![CDATA[ */ window.gtranslateSettings = /* document.write */ window.gtranslateSettings || {};window.gtranslateSettings['71620520'] = {"default_language":"en","languages":["en","fr","de","pt","es"],"url_structure":"none","flag_style":"2d","flag_size":16,"alt_flags":{"pt":"brazil"},"flags_location":"\/wp-content\/plugins\/gtranslate\/flags\/"}; /* ]]> */ </script><script src="https://odsc.com/wp-content/plugins/gtranslate/js/base.js" data-no-optimize="1" data-no-minify="1" data-gt-orig-url="/california/odsc-west-schedule-2024/" data-gt-orig-domain="odsc.com" data-gt-widget-id="71620520" defer></script><!-- start Simple Custom CSS and JS --> <script type="text/javascript"> const chartData = [ { "Title": "September 30th : Network Analysis Make Simple", "URL": "www.odsc.com/california", "Level": "Pre-Bootcamp", "Level 1": "Live", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "October 6th : NLP Fundamentals", "URL": "www.odsc.com/california", "Level": "Pre-Bootcamp", "Level 1": "Live", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "October 12th : Introduction to Python Programming", "URL": "www.odsc.com/california", "Level": "Pre-Bootcamp", "Level 1": "Live", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "October 21st : Getting Started with Practical MLOps", "URL": "www.odsc.com/california", "Level": "Pre-Bootcamp", "Level 1": "Live", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "October 26th: Boocamp Warmup : Data Literacy", "URL": "www.odsc.com/california", "Level": "Pre-Bootcamp", "Level 1": "Live", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Tutorial", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "October 28th : Gradient Boosting", "URL": "www.odsc.com/california", "Level": "Pre-Bootcamp", "Level 1": "Live", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "October 26th: Boocamp Warmup : AI Literacy", "URL": "www.odsc.com/california", "Level": "Pre-Bootcamp", "Level 1": "Live", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Tutorial", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Complete Python Fundamentals", "URL": "www.odsc.com/california", "Level": "Pre-Bootcamp", "Level 1": "On-demand", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Modern Data Acquisition using Python", "URL": "www.odsc.com/california", "Level": "Pre-Bootcamp", "Level 1": "On-demand", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Machine Learning 101", "URL": "www.odsc.com/california", "Level": "Pre-Bootcamp", "Level 1": "On-demand", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "SQL for Data Science", "URL": "www.odsc.com/california", "Level": "Pre-Bootcamp", "Level 1": "On-demand", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Recommendation Systems in Python", "URL": "www.odsc.com/california", "Level": "Pre-Bootcamp", "Level 1": "On-demand", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Machine Learning Foundations: Linear Algebra", "URL": "www.odsc.com/california", "Level": "Pre-Bootcamp", "Level 1": "On-demand", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Mathematics and Statistics for Machine Learning", "URL": "www.odsc.com/california", "Level": "Bootcamp", "Level 1": "Day 1", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Introduction to SQL", "URL": "www.odsc.com/california", "Level": "Bootcamp", "Level 1": "Day 1", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Introduction to Python", "URL": "www.odsc.com/california", "Level": "Bootcamp", "Level 1": "Day 1", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Introduction to Data Analysis", "URL": "www.odsc.com/california", "Level": "Bootcamp", "Level 1": "Day 1", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Introduction to Machine Learning Using scikit-learn", "URL": "www.odsc.com/california", "Level": "Bootcamp", "Level 1": "Day 1", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Introduction to Deep Learning", "URL": "www.odsc.com/california", "Level": "Bootcamp", "Level 1": "Day 1", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Introduction to Intermediate Deep Learning with Tensorflow", "URL": "www.odsc.com/california", "Level": "Bootcamp", "Level 1": "Day 2", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Deep Learning with Keras", "URL": "www.odsc.com/california", "Level": "Bootcamp", "Level 1": "Day 2", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Transfer Learning in NLP", "URL": "www.odsc.com/california", "Level": "Bootcamp", "Level 1": "Day 2", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "MLOps with Kubeflow & Kubernetes", "URL": "www.odsc.com/california", "Level": "Bootcamp", "Level 1": "Day 2", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Causal Reinforcement Learning", "URL": "www.odsc.com/california", "Level": "Bootcamp", "Level 1": "Day 2", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Reinforcement Learning with TF-Agents", "URL": "www.odsc.com/california", "Level": "Bootcamp", "Level 1": "Day 2", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Security and Privacy in Machine Learning", "URL": "www.odsc.com/california", "Level": "Bootcamp", "Level 1": "Day 2", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Building Machine Learning Pipelines", "URL": "www.odsc.com/california", "Level": "Bootcamp", "Level 1": "Day 2", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Advanced Deep Learning with Tensorflow", "URL": "www.odsc.com/california", "Level": "Bootcamp", "Level 1": "Day 3", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Reinforcement Learning for Human Language", "URL": "www.odsc.com/california", "Level": "Bootcamp", "Level 1": "Day 3", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "NLP Word Embeddings", "URL": "www.odsc.com/california", "Level": "Bootcamp", "Level 1": "Day 3", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Deep Learning for Detecting DeepFakes", "URL": "www.odsc.com/california", "Level": "Bootcamp", "Level 1": "Day 3", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Machine Learning at Scale using Apache Spark", "URL": "www.odsc.com/california", "Level": "Bootcamp", "Level 1": "Day 3", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Augmented & Decision Analytics", "URL": "www.odsc.com/california", "Level": "Bootcamp", "Level 1": "Day 3", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Data wrangingling", "URL": "www.odsc.com/california", "Level": "Post-Bootcamp", "Level 1": "On-demand", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "SQL", "URL": "www.odsc.com/california", "Level": "Post-Bootcamp", "Level 1": "On-demand", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Programming", "URL": "www.odsc.com/california", "Level": "Post-Bootcamp", "Level 1": "On-demand", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Supervised Learning", "URL": "www.odsc.com/california", "Level": "Post-Bootcamp", "Level 1": "On-demand", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Unsupervised Learning", "URL": "www.odsc.com/california", "Level": "Post-Bootcamp", "Level 1": "On-demand", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Reinforcement Learning", "URL": "www.odsc.com/california", "Level": "Post-Bootcamp", "Level 1": "On-demand", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Deep Learning", "URL": "www.odsc.com/california", "Level": "Post-Bootcamp", "Level 1": "On-demand", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "NLP", "URL": "www.odsc.com/california", "Level": "Post-Bootcamp", "Level 1": "On-demand", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Tools", "URL": "www.odsc.com/california", "Level": "Post-Bootcamp", "Level 1": "On-demand", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" }, { "Title": "Frameworks", "URL": "www.odsc.com/california", "Level": "Post-Bootcamp", "Level 1": "On-demand", "Level 2": "", "Level 3": "", "Level 4": "", "Level 5": "", "Type": "Training", "Duration": "90 Minutes", "Course Link": "www.someurl.com", "": "" } ]; function convertData(data) { let result = {}; let colors = ['cadetblue', 'cornflowerblue', 'darkseagreen', 'greenyellow', 'forestgreen', 'lightseagreen', 'orangered', 'mediumvioletred', 'yellow', 'saddlebrown', 'rosybrown', 'dimgray']; for (let item of data) { const level = item["Level"].trim(); const level1 = item["Level 1"].trim(); const level2 = item["Level 2"].trim(); const level3 = item["Level 3"].trim(); const level4 = item["Level 4"].trim(); const level5 = item["Level 5"].trim(); if (level && item.Title) { if (!result[level]) { result[level] = {}; } if (!level1) { result[level][item.Title] = item; continue; } if (!result[level][level1]) { result[level][level1] = {}; } if (!level2) { result[level][level1][item.Title] = item; continue; } if (!result[level][level1][level2]) { result[level][level1][level2] = {}; } if (!level3) { result[level][level1][level2][item.Title] = item; continue; } if (!result[level][level1][level2][level3]) { result[level][level1][level2][level3] = {}; } if (!level4) { result[level][level1][level2][level3][item.Title] = item; continue; } if (!result[level][level1][level2][level3][level4]) { result[level][level1][level2][level3][level4] = {}; } if (!level5) { result[level][level1][level2][level3][level4][item.Title] = item; continue; } result[level][level1][level2][level3][level4][level5][item.Title] = item; } } function convertChildrenAsArray(name, item, level, color) { if (item['Level']) { return { ...item, name: item['Title'], color, }; } let children = []; let index = 0; for (let key in item) { children.push(convertChildrenAsArray(key, item[key], level + 1, level === 0 ? colors[index] : color)); index ++; } return { name, children, color, }; } return convertChildrenAsArray("ODSC Bootcamp", result, 0, null); } const treeData = convertData(chartData); function drawChart(data) { d3.selectAll('.tooltip').remove(); d3.selectAll('#tree-container svg').remove(); const viewerWidth = window.innerWidth * 0.8; const viewerHeight = window.innerHeight - 16; const duration = 750; const levels = 6; const maxTextWith = 150; const lineHeight = 14; let hoveredPosition = { x: 0, y: 0 }; // Define the div for the tooltip let tooltip = d3.select("body").append("div") .attr("class", "tooltip") .style("opacity", 0); let nodeId = 0; let root; let tree = d3.layout.tree() .size([viewerHeight, viewerWidth]); // define a d3 diagonal projection for use by the node paths later on. const diagonal = d3.svg.diagonal() .projection(d => [d.y, d.x]); // Define the zoom function for the zoomable tree function zoom() { svgGroup.attr("transform", "translate(" + d3.event.translate + ")scale(" + d3.event.scale + ")"); } // define the zoomListener which calls the zoom function on the "zoom" event constrained within the scaleExtents const zoomListener = d3.behavior.zoom().scaleExtent([0.1, 3]).on("zoom", zoom); // define the baseSvg, attaching a class for styling and the zoomListener const baseSvg = d3.select("#tree-container").append("svg") .attr("width", viewerWidth) .attr("height", viewerHeight) .attr("class", "overlay") .on("mousemove", (d) => mouseOut(d)); // Function to collapse node function collapse(d) { if (d.children) { d._children = d.children; d._children.forEach(collapse); d.children = null; } } // Function to center node when clicked/dropped so node doesn't get lost when collapsing/moving with large amount of children. function centerNode(source) { let scale = zoomListener.scale(); let x = -source.y0; let y = -source.x0; x = x * scale + viewerWidth / 2; y = y * scale + viewerHeight / 2; d3.select('g').transition() .duration(duration) .attr("transform", "translate(" + x + "," + y + ")scale(" + scale + ")"); zoomListener.scale(scale); zoomListener.translate([x, y]); } // Toggle children function function toggleChildren(d) { if (d.children) { d._children = d.children; d.children = null; } else if (d._children) { d.children = d._children; d._children = null; } return d; } // Toggle children on click. function click(d) { if (d3.event.defaultPrevented) { return; } d = toggleChildren(d); update(d); if (d.children || d._children) { centerNode(d); } if (d.depth === 0) { if (d.children) { baseSvg.call(zoomListener); } else { baseSvg.on('.zoom', null); } } } // Get width of text function measureWidth(text) { const context = document.createElement("canvas").getContext("2d"); return context.measureText(text).width; } // Split string into words function splitWords(text, node) { const words = text.split(/\s+/g); // To hyphenate: /\s+|(?<=-)/ if (!words[words.length - 1]) { words.pop(); } else if (!words[0]) { words.shift(); } return { words, node }; } // Get lines from text when it overflows maxTextWidth function getLines(node) { let line; let lineWidth0 = Infinity; let lines = []; for (let i = 0, n = node.words.length; i < n; ++i) { let lineText1 = (line ? line.text + " " : "") + node.words[i]; let lineWidth1 = measureWidth(lineText1); if ((lineWidth0 + lineWidth1) / 2 < maxTextWith) { line.width = lineWidth0 = lineWidth1; line.text = lineText1; } else { lineWidth0 = measureWidth(node.words[i]); line = { width: lineWidth0, text: node.words[i], ...node.node, length: lines.length }; lines.push(line); } } return lines; } function mouseOver(d) { if (!d["Level"]) { return; } let html = ""; if (d["URL"]) { html += `<a target="_blank" rel="noopener">Course Link</a>`; } html += `<p><strong>Level: </strong>${d["Level"]}</p>`; if (d["Type"]) { html += `<p><strong>Type: </strong>${d["Type"]}</p>`; } if (d["Duration"]) { html += `<p><strong>Duration: </strong>${d["Duration"]}</p>`; } tooltip.transition() .duration(200) .style("opacity", .9); tooltip.html(html) .style("left", (d3.event.pageX) + "px") .style("top", (d3.event.pageY) + "px"); hoveredPosition = { x: d3.event.pageX, y: d3.event.pageY }; } function mouseOut(d) { if (hoveredPosition.x === 0 || (d3.event.pageX > hoveredPosition.x - 10 && d3.event.pageX < hoveredPosition.x + 150 && d3.event.pageY > hoveredPosition.y - 10 && d3.event.pageY < hoveredPosition.y + 90)) { return; } tooltip.transition() .duration(500) .style("opacity", 0); tooltip.html(null); } function update(source) { // Compute the new height, function counts total children of root node and sets tree height accordingly. // This prevents the layout looking squashed when new nodes are made visible or looking sparse when nodes are removed // This makes the layout more consistent. let levelWidth = [1]; const childCount = (level, n) => { if (n.children && n.children.length > 0) { if (levelWidth.length <= level + 1) { levelWidth.push(0); } levelWidth[level + 1] += n.children.length; n.children.forEach((d) => { childCount(level + 1, d); }); } }; childCount(0, root); const newHeight = d3.max(levelWidth) * (25 + (source.depth ? source.depth * 5 : 0)); // 25 pixels per line tree = tree.size([newHeight, viewerWidth]); // Compute the new tree layout. const nodes = tree.nodes(root).reverse(), links = tree.links(nodes); // Set widths between levels based on maxLabelLength. nodes.forEach((d) => { d.y = (d.depth * 300); d.x = d.x * 2.5; }); // Update the nodes… node = svgGroup.selectAll("g.node") .data(nodes, (d) => { return d.id || (d.id = ++nodeId); }); // Enter any new nodes at the parent's previous position. let nodeEnter = node.enter().append("g") .attr("class", "node") .attr("transform", (d) => { return "translate(" + source.y0 + "," + source.x0 + ")"; }) .on('click', click); nodeEnter.append("circle") .attr('class', 'nodeCircle') .attr("r", 0) .style("fill", (d) => { return d._children ? (d.color || "lightsteelblue") : "#fff"; }); let rect = nodeEnter.append("rect") .attr('class', 'nodeRect') .attr('x', (d) => 5) .attr('y', (d) => { return d.children || d._children ? 0 : -12 - getLines(splitWords(d.name, d)).length * 2; }) .attr('width', (d) => { return d.children || d._children ? 0 : 240; }) .attr('height', (d) => { return d.children || d._children ? 0 : getLines(splitWords(d.name, d)).length * 18; }) .attr('stroke', '#fff') .attr('fill', 'transparent') .attr('rx', 10); nodeEnter.append("text") .attr('class', 'nodeText') .attr("text-anchor", (d) => { return d.children || d._children ? "end" : "start"; }) .style("fill", (d) => { return d.children || d._children ? "#ddd" : "#fff"; }) .style("fill-opacity", 0) .on("mouseover", (d) => mouseOver(d)) .selectAll("tspan") .data(d => getLines(splitWords(d.name, d))) .enter() .append("tspan") .attr("x", (d) => { return d.children || d._children ? -8 : 8; }) .attr("y", (d, i) => i * lineHeight) .attr("dy", d => { if (d.children || d._children) { return '-1.2em'; } return '0em'; }) .text(d => d.text); // Change the circle fill depending on whether it has children and is collapsed node.select("circle.nodeCircle") .attr("r", d => { let maxChildLength = 0; if (d.children || d._children) { (d.children || d._children).forEach(c => { if ((c.children || c._children) && (c.children || c._children).length > maxChildLength) { maxChildLength = (c.children || c._children).length; } }); if ((d.children || d._children).length > maxChildLength) { maxChildLength = (d.children || d._children).length; } if (maxChildLength < 4) { maxChildLength = 4; } } else { maxChildLength = 1.5; } return (maxChildLength * 3 / 2) + "px"; }) .style("fill", (d) => { return d._children ? (d.color || "green") : "#fff"; }); // Transition nodes to their new position. let nodeUpdate = node.transition() .duration(duration) .attr("transform", (d) => { return "translate(" + d.y + "," + d.x + ")"; }); // Fade the text in nodeUpdate.select("text") .style("fill-opacity", 1); // Transition exiting nodes to the parent's new position. let nodeExit = node.exit().transition() .duration(duration) .attr("transform", (d) => { return "translate(" + source.y + "," + source.x + ")"; }) .remove(); nodeExit.select("circle") .attr("r", 0); nodeExit.select("text") .style("fill-opacity", 0); // Update the links… let link = svgGroup.selectAll("path.link") .data(links, (d) => { return d.target.id; }); // Enter any new links at the parent's previous position. link.enter().insert("path", "g") .attr("class", "link") .attr("stroke-width", d => { if (d.target.children || d.target._children) { return (d.target.children || d.target._children).length * 3 + "px"; } return "2px"; }) .attr("stroke", d => d.target.color || "#ccc") .attr("d", (d) => { const o = { x: source.x0, y: source.y0 }; return diagonal({ source: o, target: o }); }); // Transition links to their new position. link.transition() .duration(duration) .attr("d", diagonal); // Transition exiting nodes to the parent's new position. link.exit().transition() .duration(duration) .attr("d", (d) => { const o = { x: source.x, y: source.y }; return diagonal({ source: o, target: o }); }) .remove(); // Stash the old positions for transition. nodes.forEach((d) => { d.x0 = d.x; d.y0 = d.y; }); } // Append a group which holds all nodes and which the zoom Listener can act upon. let svgGroup = baseSvg.append("g"); // Define the root root = JSON.parse(JSON.stringify(data)); root.x0 = viewerHeight / 2; root.y0 = 0; // Layout the tree initially and center on the root node. collapse(root); update(root); centerNode(root); } drawChart(treeData); window.addEventListener('resize', () => { drawChart(treeData); });</script> <!-- end Simple Custom CSS and JS --><div id='icon_wrapper'><a target="_blank" class='fuse_social_icons_links' href='https://www.facebook.com/OPENDATASCI'> <i class='fa fa-facebook fb-awesome-social awesome-social'></i></a><br /><a target="_blank" class='fuse_social_icons_links' href='https://twitter.com/_odsc'> <i class='fa-brands fa-x-twitter tw-awesome-social awesome-social'></i></a><br /><a target="_blank" class='fuse_social_icons_links' href='https://www.linkedin.com/company/open-data-science/'> <i class='fa fa-linkedin linkedin-awesome-social awesome-social'></i></a><br /><a target="_blank" class='fuse_social_icons_links' href='https://www.youtube.com/c/OpenDataScienceCon'> <i class='fa fa-youtube youtube-awesome-social awesome-social'></i></a><br /><a target="_blank" class='fuse_social_icons_links' href='https://instagram.com/odsc/'> <i class='fa fa-instagram instagram-awesome-social awesome-social'></i></a><br /><a target="_blank" class='fuse_social_icons_links' href='https://bit.ly/35pfPZo'> <i class='fa fa-slack slack-awesome-social awesome-social'></i></a><br /><a target="_blank" class='fuse_social_icons_links' href='mailto:info@odsc.com'> <i class='fa fa-envelope envelope-awesome-social awesome-social'></i></a><br /> <!-- Cookie Notice plugin v2.4.18 by Hu-manity.co https://hu-manity.co/ --><div id="cookie-notice" role="dialog" class="cookie-notice-hidden cookie-revoke-hidden cn-position-bottom" aria-label="Cookie Notice" style="background-color: rgba(50,50,58,1);"><div class="cookie-notice-container" style="color: #fff"><span id="cn-notice-text" class="cn-text-container">We use cookies to ensure that we give you the best experience on our website. If you continue to use this site we will assume that you are happy with it.</span><span id="cn-notice-buttons" class="cn-buttons-container"><a href="#" id="cn-accept-cookie" data-cookie-set="accept" class="cn-set-cookie cn-button" aria-label="Confirm" style="background-color: #00a99d">Confirm</a><a href="#" id="cn-refuse-cookie" data-cookie-set="refuse" class="cn-set-cookie cn-button" aria-label="No" style="background-color: #00a99d">No</a><a href="https://odsc.com/privacy-policy/" target="_blank" id="cn-more-info" class="cn-more-info cn-button" aria-label="Privacy policy" style="background-color: #00a99d">Privacy policy</a></span><span id="cn-close-notice" data-cookie-set="accept" class="cn-close-icon" title="No"></span></div><div class="cookie-revoke-container" style="color: #fff"><span id="cn-revoke-text" class="cn-text-container">You can revoke your consent any time using the Revoke consent button.</span><span id="cn-revoke-buttons" class="cn-buttons-container"><a href="#" class="cn-revoke-cookie cn-button" aria-label="Revoke consent" style="background-color: #00a99d">Revoke consent</a></span></div></div> <!-- / Cookie Notice plugin --><!-- Option Panel Custom JavaScript --> <script> //jQuery(document).ready(function(){ jQuery(document).ready(function($){ function timeout() { setTimeout(function () { console.log('fired'); if(jQuery.fn.hasOwnProperty('magnificPopup')){ jQuery('.mptt-shortcode-wrapper').magnificPopup({ delegate: 'a', // child items selector, by clicking on it popup will open type: 'ajax', callbacks: { parseAjax: function(mfpResponse) { var title = $(mfpResponse.data).find('.post-title a').text(); var content = $(mfpResponse.data).find('.post-content').html(); mfpResponse.data ="<h4>"+title +'</h4>'+ content; } } }); } else{ timeout(); } }, 500); } if(jQuery('.mptt-shortcode-wrapper').length){ timeout(); } // jQuery('.mptt-shortcode-table:not(:first)').remove(); }); // }); </script> <!-- Start of HubSpot Embed Code --> <script type="text/javascript" id="hs-script-loader" async defer src="//js.hs-scripts.com/1865444.js"></script> <!-- End of HubSpot Embed Code --> <script type="text/javascript"> adroll_adv_id = "MN7XH5FMX5AODAUTQ7LEYH"; adroll_pix_id = "4WF3QPQPGRHLJNYPUIJSDC"; /* OPTIONAL: provide email to improve user identification */ /* adroll_email = "username@example.com"; */ (function () { var _onload = function(){ if (document.readyState && !/loaded|complete/.test(document.readyState)){setTimeout(_onload, 10);return} if (!window.__adroll_loaded){__adroll_loaded=true;setTimeout(_onload, 50);return} var scr = document.createElement("script"); var host = (("https:" == document.location.protocol) ? "https://s.adroll.com" : "http://a.adroll.com"); scr.setAttribute('async', 'true'); scr.type = "text/javascript"; scr.src = host + "/j/roundtrip.js"; ((document.getElementsByTagName('head') || [null])[0] || document.getElementsByTagName('script')[0].parentNode).appendChild(scr); }; if (window.addEventListener) {window.addEventListener('load', _onload, false);} else {window.attachEvent('onload', _onload)} }()); </script> <!-- FAQ --> <script type="text/javascript"> $(".parent").click(function(){ $(".parent").toggleClass("activated"); $(".faq_categories").toggleClass("activated"); $(".subcategories").toggleClass("activated"); }); </script></body></html>