Microsoft Certified: Azure AI Apps and Agents Developer Associate

Your team is instrumenting a generative AI application and wants to correlate every piece of telemetry produced by a single user interaction, even when the request crosses multiple services. Which tagging approach should you implement to achieve this?

During a review of latency, stakeholders ask how you plan to represent the range of response experiences that different users might have with your generative AI service. Which reporting approach aligns with the recommended practice?