LLM-Evals6 min read
How to build an evaluation dataset from your production traces
Build a golden dataset from production traces in LangWatch: search traces, judge every conversation with Instant Evals, query with LangWatch SQL, add to a dataset and automate it.
By Rogerio ChavesLangWatch Blog

Read the full story
Originally published by LangWatch Blog on Nov 4, 2026.
Comments
Sign in to join the discussion.