AWS · AIP-C01
Monitoring generative AI applications
AIP-C01 · Lesson 20 of 23 · 2:41:23
YouTube loads only after you press play. This chapter opens at its timestamp in the full course. Watch on YouTube ↗
In this lesson
Prepare for the AIP-C01 exam (AWS Certified Generative AI Developer - Professional). In this lesson: holistic observability with CloudWatch generative AI observability, X-Ray, and business dashboards, generative AI metrics and model invocation logging, anomaly detection, tool and agent observability with AgentCore Observability, vector store operations, and generative AI failure modes. Includes three original practice questions.
Preview the topics
- AnyCompany wants to review the exact prompts and responses behind last week’s weak answers, but finds no trace of them. What should it have enabled?
- Token usage on AnyCompany’s support agent follows a daily pattern, and the team wants an alarm only when it leaves that pattern, without a fixed threshold. What should it use?
- After a model upgrade, AnyCompany wants to know whether the answers to the same two hundred reference questions changed, and which ones. Which technique fits?