LangChain details how its Labs team trains custom models for LangSmith Engine with Baseten Loops

LangChain · x · 2026-09-16

LangChain shares how its Labs team uses Baseten Loops to train custom models for LangSmith Engine, its in-platform agent. The workflow: pick a problem/agent to improve, curate data and evals/environments, run many training experiments with continuous measurement, and loop. The thesis: the future is continuously turning valuable data into better/faster/cheaper models — 'Owning your Intelligence' — and today's infra lets any team run hundreds of experiments to optimize quality/cost/latency.

Related event: LangChain Trains Custom Agent Models on Baseten Loops(2 posts)→

Original post →

More from coding & agent

coding & agent channel →