AI Observability
Use Trace Explorer
Find AI requests, inspect trace groups, and debug errors or slow responses.
Trace workflow
- Open AI Observability > Traces.
- Filter by time range, provider, model, service, environment, or status.
- Open a trace group to see related requests.
- Open request detail to inspect spans, timing, token usage, metadata, and error text.
- Use Errors, Cost, or Models when one trace points to a larger pattern.
If traces are empty
- Confirm the ingest key is active.
- Confirm the app is sending to the correct backend URL.
- Clear filters and widen the time range.
- Check whether the app sends telemetry only on successful requests.
Need more help?
Keep moving with the right next guide
If your team still runs into setup issues, empty dashboards, billing delays, callback failures, or alerting problems, continue with troubleshooting before re-running the entire onboarding flow.