July 27, 2026
Review traces with AI evaluators
Every trace can now be reviewed automatically, not just the ones your team has time to open.
Updates
- Write detections in codeSeptember 7, 2026
- Review traces with AI evaluatorsJuly 27, 2026
- Replay an agent run step by stepJuly 1, 2026
- See cost, latency, and errors across every runJune 16, 2026
- Triage in Your EditorMay 29, 2026
- Import Your Traces from LangSmith and BraintrustMay 28, 2026
- Ask your trace data anythingMay 1, 2026
- Keep sensitive data out of your tracesApril 12, 2026
- Version your prompts like codeMarch 5, 2026
- Instrument multi-turn conversationsJanuary 20, 2026
- Collaborate on traces with your teamDecember 10, 2025
- See exactly what your AI agent didNovember 18, 2025
- Catch issues before your users doNovember 5, 2025
- Zero-config observability for every frameworkOctober 22, 2025
How it works
Build custom AI evaluators with your own rubric and verdict schema. Use an LLM judge for a fast one-pass review, or an agent judge when the evaluation needs to inspect span data, call tools, or check external context first.
Agent judges can use connected integrations, custom MCP servers, or allowlisted HTTP fetches during review.
Test an evaluator against a real trace in the playground, then assign it per question alongside your human reviewers.
Evals now live at /evals, with old /feedback links redirecting automatically. Trace and span eval results are also exposed to agents querying over MCP, so evaluation outcomes can feed back into debugging and investigation workflows.