What Is LangChain, Actually?

Amy and Sean from LangChain explain what the company actually does today. LangChain started as an open source framework that predates ChatGPT, helping engineers wrangle LLMs before most people had even heard the word prompt. As developers moved toward building agents, LangChain released LangGraph, an agent runtime and low level orchestration framework. Now LangChain serves the full agent development lifecycle: building with LangChain, LangGraph, and Deep Agents, testing with evals and regression testing, deploying and monitoring through LangSmith. More than 7000 organizations, including NVIDIA, Bridgewater, and Harvey, use LangSmith to ship agents that hold up in production for customers with their brand image on the line. This is a clip from our full conversation where Amy and Sean answer the most commonly searched questions about AI agents, from the basics like what is an agent, what is RAG, and what is MCP, to harder questions like why agents hallucinate, how to evaluate them, and what LangChain actually does.