LangChain and Fireworks Workshop: Train Custom Eval Models for Better Agents

LangChain · x · 2026-08-21

LangChain and Fireworks host a hands-on SF workshop on August 25 (plus rooftop happy hour) about agent evals: human-in-the-loop evaluation is costly while LLM-as-judge can be unreliable and biased. Attendees will learn to train customized eval models usable in LangSmith to optimize agent performance, closing the loop from evals into fine-tuning. Aimed at AI/ML engineers and product builders shipping LLM or agent features.

Original post →

More from coding & agent

coding & agent channel →