LangWatch · The test-harness layer of the agent stack — the people trying to make "does this agent actually work" a number you can put in CI, priced per event in euros and open enough to run on your own metal.

A platform for testing AI agents before they ship and watching them after. An LLM plays the user and holds a multi-turn conversation with your agent — in text or in voice — a judge scores the transcript against criteria you wrote in plain English, and the same scenarios run on your laptop and in CI. Around that sit production tracing, evaluations, prompt versioning, and a gateway that routes and budgets model spend. Open source (Apache 2.0) and self-hostable.

Sector

Evals & observability

Stage

Pre-Seed

On the roster since

15 Sept 2026