Video is one of the richest data types we have, but most systems still treat it like static text. In SF, Qdrant and TwelveLabs are hosting a live technical session to unpack how to build semantic video recommendation systems using multimodal embeddings, real vector infrastructure, and open-source tools. We’ll walk through the stack and a working GitHub project you can build on. RSVP if you are in the Bay Area: https://lu.ma/26jxsi5x