LangChain plans Nov. 5 San Francisco meetup on decision models and AI-agent evaluation

LangChain is promoting a Nov. 5, 2026, meetup in San Francisco with TypeSafe AI about decision models and evaluating AI agents. Called “LangChain SF Meetup: Judgment Calls,” the event is scheduled for 6 to 8:30 p.m. PST.

The agenda lists three talks. Sydney Runkle, a LangChain open-source product manager, is set to discuss pairing decision models with LLMs in LangGraph, using a “cheap by default, frontier on exception” pattern. Allie Laabs, TypeSafe AI’s founding developer advocate, plans to argue for thinking of Jev as a “System One Model.” LangChain education engineer Seán Roche is scheduled to discuss Jev as an evaluator of AI agents.