Sentry enforces. Observatory explains. Together they give you a signed, queryable record of every AI decision your business has ever made.
In-line AI governance. Screens every prompt against your policies before it reaches a model provider.
Post-hoc observability, tracing, and drift-detection across every AI decision.
# 1. Point your SDK to the Sentry Gateway export OPENAI_API_BASE="https://gateway.sentry.lunar/v1" # 2. Every request is now screened, logged, and provable openai.ChatCompletion.create(model="gpt-4", messages=[...])