Loading video player...
Captured at the Austin LangChain / AI Middleware Users Group (AIMUG) November meetup — four practical talks and demos for builders who want to ship reliable AI systems. Speakers & highlights: Colin McNamara — Community update + Intro to LangGraph: controllability, persistence, streaming, human-in-the-loop, and how LangGraph fits into the LangChain ecosystem. Anupama Garani — RAG in the Enterprise: why many enterprise RAG projects fail, the role of state/context/memory, and a demo showing stateful RAG with LangGraph. Collier King — Deep Agents: an experiment doing "deep research" across 400 companies using subagents, middleware patterns, lessons on prompts, logging, cost & token tradeoffs. Paul Phelps (remote) — AI Project Management: why traditional PM playbooks fail for AI, framework comparisons, organizational red flags, and practical tips to increase the odds of getting AI into production. Why watch Real-world tactics for building reliable RAG and agentic systems. Hands-on lessons about observability, persistence, middleware, and cost control. Actionable PM & org guidance so your models actually ship. Chapters (timestamps are approximate) 00:00 — Welcome & AIMUG community notes (Colin McNamara) 05:30 — Intro to LangGraph: control flow, persistence, streaming, LangSmith integration (Colin) 36:20 — RAG in Enterprise: failure zones, state vs. context vs. memory, travel-agent demo (Anupama Garani) 58:50 — Deep Agents: subagents, middleware, structured outputs, cost & token lessons (Collier King) 1:44:30 — AI Project Management: frameworks, red flags, getting AI into production (Paul Phelps) End — Q&A, community announcements & after-party info Resources & links AIMUG: https://aimug.org — join the community, office hours, and slides. Search YouTube: "AIMUG" for past meetup videos and resources. If you found this helpful — like, subscribe, and drop a comment with: Which talk you’d like a deeper tutorial on (LangGraph, RAG, Deep Agents, middleware, or PM frameworks).