A practical guide series for learning LangSmith - the platform that lets you see exactly what your LLM app did, test it against real examples instead of vibes, and fix a bad prompt without redeploying. Written by people who build agents for a living.
One request in your app becomes one trace in LangSmith - every LLM call, tool call, and retriever step nested inside it.
Never opened LangSmith, or opened it once and got lost? These two guides get you oriented.
The plain-English version: what problem it solves, how it relates to LangChain and LangGraph, and why "it worked when I tested it" isn't good enough.
→Three ideas that cover almost everything - once these click, the whole dashboard makes sense.
Time to see your own app's traces show up, not just read about it.
APA Mastery runs live, practical sessions on working with modern AI tools - not just theory.
See What's On →The two things that separate "I glanced at some traces" from an app you'd actually trust in production.
Turning real traces into a test suite, and how LLM-as-judge evaluators score things a plain assertion can't.
→Iterating on prompts without redeploying code, and how versioned commits keep production from breaking mid-edit.