A practical guide series for learning LangChain - prompts, models, retrieval, tools, and exactly where LangChain's job ends and LangGraph's begins. Written by people who build with it for a living.
LangChain's core pattern: pipe simple pieces together with | to build a chain.
New to LangChain, or coming from the LangGraph guides and wondering how the two fit together? Start here.
The toolbox versus the wiring diagram - what LangChain actually gives you, and exactly where it hands off to LangGraph.
→The three pieces you'll chain together in almost every LangChain project, and how LCEL's pipe syntax connects them.
Where LangChain earns its keep - connecting a model to documents it wasn't trained on.
APA Mastery runs live, practical sessions on working with modern AI tools - not just theory.
See What's On →The two areas where LangChain's own guidance has shifted the most recently - worth getting the current picture, not an outdated one.
The mechanics underneath everything above, plus how to actually know if a change made your chain better.
What the | operator actually does, and why LCEL chains get streaming and async for free.
The step before retrieval: getting documents in, and splitting them into chunks that actually retrieve well.
→Getting a model to reliably return JSON matching a schema, and when that beats reaching for a full agent.
→How to tell whether a prompt or chunking change actually made your chain better, instead of guessing.