LangChain SF Meetup: LLM Wikis and Giving Your Agents Memory

Join us for an meetup in San Francisco with Brace Sproul (Head of Applied AI at LangChain), and Jacob Teo (Research & Engineering at Cognition).

Most agents fail not because the model is weak, but because it doesn't have the right context.

LLM wikis are emerging as a common pattern for solving this: a structured knowledge base an agent can query to understand a codebase, product, or domain. LangChain recently open sourced OpenWiki, a project for building and maintaining your own LLM wiki.

This meetup will feature two talks. First, Brace will break down how to actually build these systems, and whether "wiki" is even the right way to think about them. Then, Jacob will share more on Cognition's DeepWiki.

We'll cover:

What is an LLM wiki?

Are LLM wikis the right abstraction for agent context?

Are LLM wikis built for agents, for humans, or both?

What's the hardest part of actually building one?

How does Cognition's DeepWiki actually work?

We'll also get into where this pattern is headed and discuss what it takes to keep an LLM wiki useful as a codebase or product evolves.

Agenda ⏰

6:00 PM: Welcome + Food/Drinks

6:30 PM:  Presentation with Brace Sproul (LangChain) - LLM Wikis and Giving Your Agents Memory with OpenWiki

6:50 PM: Q&A Session with Brace

7:00 PM: Presentation with Jacob Teo (Cognition) - Cognition's DeepWiki

7:15 PM: Q&A Session with Jacob

7:25 PM: Networking

8:30 PM: Event Ends

Event info:

🎤 Please note this event is fully in-person and will not be live-streamed.

📍 Location: SOMA — address provided upon approval.

🎟️ We can only admit guests with approved registrations.

About the host:

LangChain powers the full agent development lifecycle — building, testing, deploying, and monitoring — so AI teams can improve their agents systematically. LangSmith Engine accelerates this cycle, automatically surfacing and fixing issues to improve agents over time. LangSmith is neutral by design, so teams can customize their own stack to optimize on cost and performance as the landscape evolves. More than 7,000 customers, including Nvidia, Bridgewater, LinkedIn, Workday, Harvey, and Rippling trust LangSmith to build and manage their agents. Learn more: www.langchain.com