Hey there,
Welcome to Opik, glad you signed up!
I'm Jacques, Head of Product. Opik is an observability and eval platform for agents — so you can debug, test, and improve your agent with confidence at every stage. Get started in under 2 minutes with our
Quickstart Guide →
Three things Opik does for you:
- Observability: See what your agent is doing. Every step, tool call, and reasoning trace.
- Evals: Know if your agent really works. Pass/fail assertions or quantitative eval metrics, your call.
- Continuous Improvement: Track quality, cost, and latency in production. Get alerted when something breaks and fix it fast with AI-powered suggestions.
Ready to instrument in two minutes?
- In Claude Code, Cursor, Codex, or whatever you use, run npx skills add comet-ml/opik-skills
- Then prompt: Instrument my agent with Opik using the /instrument command
- Run your agent. Your traces will show up in Opik.
Not using a coding agent? Choose your framework or model provider to
integrate here→
Jacques
Head of Product