Question

Your team has deployed a customer-support agent to production in Azure AI Foundry. Leadership is concerned about unpredictable monthly spend and wants a dashboard that shows how many tokens each conversation consumes over time, without writing custom instrumentation code. Which Foundry capability should you enable to satisfy this requirement?

Accepted Answer

Enable continuous monitoring in Foundry and use the built-in cost/token consumption metrics on the application's monitoring dashboard

Answer

Configure a groundedness evaluation on a scheduled batch run to measure response quality

Answer

Add a custom metric in a prompt flow evaluation dataset that counts characters in each response

Answer

Enable Azure Content Safety filters on the serverless endpoint to block harmful outputs