Question

Your team runs a customer-facing chat agent deployed in Azure AI Foundry. Users complain that responses occasionally feel slow, but the overall average response time on your dashboard looks acceptable. Leadership wants to confirm that the agent meets a service-level objective stating that 95% of requests must complete within 3 seconds. Which observability approach in Foundry best validates whether this SLO is being met?

Accepted Answer

Track the P95 latency percentile metric and alert when the 95th-percentile response time exceeds 3 seconds.

Answer

Monitor the mean (average) latency metric over a rolling 24-hour window and alert when it exceeds 3 seconds.

Answer

Monitor total token consumption per request and correlate it with cost to infer whether latency is acceptable.

Answer

Enable groundedness and relevance evaluations on production traffic to detect the source of slow responses.