Bay Area meetup on managing AI agent memory and context
Original: If you have had success with Gbrain/LLM Wiki/other context management techniques for agents, and you're in the Bay Area, would love to invite you to demo at this event!
https://luma.com/rey4v8r8 ↗Deep summary
What's new: Zara Zhang (Radical Ventures, AI investor) is organizing a Demo Day in San Francisco focused on context management for AI agents, co-hosted with Jules Qiu (Radical Ventures) and Kyle Atkinson, with Notion as a venue/co-host. The event targets builders who have working solutions to the management problem in agentic systems, specifically techniques like and LLM Wiki patterns.
How it works: The format is rapid-fire 5-minute live demos with Q&A, screen-shares of real workflows only, no slides. Attendance requires registration approval and token ownership verification via a crypto wallet, suggesting a web3-gated guest list. The event is explicitly practitioner-facing: no theoretical talks, only demonstrated workflows that address specific failure modes in production agent systems.
Why it matters: The problems framed for the event map directly to the hardest unsolved engineering challenges in deploying agents at organizational scale. Turning unstructured tribal knowledge into queryable corpora, curating signal-to-noise in , managing costs across long-running sessions, and deciding what to surface at each agent step are all open engineering problems with no consensus solutions. The event is effectively a curated sampling of what is actually working in practice rather than what is proposed in research.
Caveats: The token-gated registration creates selection bias toward a specific community demographic. The 5-minute demo format will surface workflow snapshots rather than rigorous comparative evaluations, so claims about what "works" will be hard to generalize. No public proceedings or recordings are mentioned.