
Beyond the Frame: Multimodal Video Recommendations with TwelveLabs + Qdrant
Qdrant AI Builders: Video Recommendations with Twelve Labs
What if your app could understand what’s happening in a video — not just the title or transcript, but the emotion in the scene, the objects on screen, and the context of the conversation?
Join Qdrant and Twelve Labs for a live, behind-the-scenes presentation on building smarter video recommendation systems using state-of-the-art vector search and multimodal AI. We’ll walk through a real open-source demo that combines Twelve Labs’ video intelligence API with Qdrant’s vector database to enable rich, semantic recommendations based on what’s actually happening inside the video.
You’ll learn:
How Marengo and Pegasus foundation models are transforming video understanding
Explore real-world applications in sports, media, and security solutions
How multimodal embeddings work across audio, visual, and textual signals
How to store and search them at scale using Qdrant
What it takes to build a recommendation engine that feels intelligent
Key takeaways from the GitHub project
This event is for developers, ML engineers, product builders, and anyone curious about the next generation of video search and personalization.