Introducing LangSmith Tuned Evaluators They automatically score agent behavior i

Online evaluation of production agent traces is usually priced out by frontier-model judges, and a tuned small model that claims better accuracy at a fraction of the cost changes that calculus.

• AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
• eval — A repeatable test for AI quality — a set of tasks plus scoring — used the way software teams use test suites, because model output is too variable to judge by eyeballing.

• postLangSmith Fine-Tuning turns agent traces into fine-tuned models
• postManaged Deep Agents 0.8 adds HTTP channels for webhook-driven agents
• postLangSmith Ships LLM-as-Judge Scoring for Every Production Trace
• postLangSmith Gateway Adds Decision Models, Offers SemIf Free for a Week

• postLangSmith Fine-Tuning turns agent traces into fine-tuned modelsLangChain
• postHow to really stop your agents from making the same mistakesGarry Tan
• articleAutonomous Agent Improvement with LangSmith EngineLangChain

Checking sign-in…

Loading comments…