LangChain · AI

Catch Agent Regressions Before You Ship: Evals for Managed Deep Agents

Once you've built a managed agent, you need a way to evaluate its performance over time, so that adding skills, tools, and other changes doesn't quietly break what already worked. Nathan from the LangChain product team walks through the eval capabilities built into Managed Deep Agents: how Harbor runs each eval in a fresh container, how an eval breaks down into an environment, a job, and a check, and how to scaffold evals with mda evals init, han

Introductie van de bron.