Back

Observability & Evals

Articles and updates grouped under this category.

Showing 1 - 3 of 3 Posts

Your Prompts Are Probably Broken. You Just Don't Have the Data to Prove It.

Your Prompts Are Probably Broken. You Just Don't Have the Data to Prove It.

Stop guessing at prompt quality. GEPA evolves your system prompts automatically — real production data, multi-metric scoring, full iteration audit.

Andrej Gamser

18 Min Read◆May, 21 2026

Observability & Evals

You Can Now Evaluate AI-Generated Video on FastRouter

FastRouter now supports AI video evaluation with LLM-as-judge scoring. Automate quality checks on Veo, Sora, and Kling — no manual review.

Ritesh Prasad

4 Min Read◆May, 7 2026

Observability & Evals

Passing Evals Aren't a Quality Signal

A high eval pass rate tells you your test set is easy, not that your system is working. A practitioner argument for adversarial evaluation, done right

Siv Souvam

8 Min Read◆April, 22 2026