TwelveLabs + Qdrant: AI Memory for Video Intelligence
AI is getting better at reasoning, but it still forgets almost everything.
Join us for an evening with engineers from TwelveLabs and Qdrant exploring what it takes to build AI systems with real memory. From making video searchable across time to giving agents long-term memory they can actually use, you'll see how modern retrieval systems are changing what's possible.
Expect practical architectures, live demos, and plenty of time for questions, food, and networking.
Agenda
5:30 PM - 6:30 PM 🍕 Food, drinks & networking
6:30 PM - 7:00 PM 🎤 Video Has No Memory. Here's How We Built One
James Reque, TwelveLabs
Every video AI interaction starts from scratch. Learn how TwelveLabs built a persistent memory layer for video, making it possible to track entities across time and query what an entire video library knows, not just individual clips.
7:00 PM - 7:30 PM
🎤 Building a Collective Memory for Edge AI
Dylan Couzon, Qdrant
Fleet memory is a shared memory layer where devices learn locally, exchange knowledge through vector search, and build a collective understanding over time. See how Qdrant Edge enables fully offline recognition, automatic synchronization, and intelligent fleets that improve with every interaction.
7:30 PM - 7:50 PM ⚡ Community Demos A series of short demos from members of the community showcasing interesting projects in AI, retrieval, agents, and video intelligence.
7:50 PM - 9:00 PM 🤝 Food, drinks & networking
About the Speakers:
- Dylan Couzon is a Developer Relations Engineer at Qdrant specializing in Retrieval systems, AI agents, robotics and embedded systems. He has spent years helping teams build production-grade AI systems that go beyond demos and hype. With a background in AdTech infrastructure at billion-device scale and a long-standing commitment to open source, he focuses on what matters in practice: retrieval quality, latency, and resilient architecture.
- James Le leads Developer Experience at TwelveLabs, where the company builds video intelligence infrastructure that enables production-grade video understanding across sports organizations, media workflows, and security applications. His work focuses on translating video AI research capabilities into deployment-ready architectures for organizations building vision-enabled intelligent systems.
Community demo slots
Have something cool to share? We're reserving a few 10-minute demo slots for community members building AI, retrieval, video intelligence, or agentic applications. If you'd like to present, reach out to [email protected] or let us know when you register.