Reference · Glossary
SLO (service level objective)
A **target for reliability or quality** — e.g. p95 latency under 2s or citation rate above 85% — that triggers alerts when breached.
When to use
Production chatbots, RAG apps, and agent APIs where you need objective “healthy vs not” signals.
When not to
Toy demos with no users — start with simple logging before formal SLOs.
Example
“99% of `/api/chat` responses complete in < 3s” — page if p95 doubles for 15 minutes.