AWS · AIF-C01
The ML lifecycle and MLOps
AIF-C01 · Lesson 4 of 17 · 23:14
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 AIF-C01 exam (AWS Certified AI Practitioner). In this lesson: the stages of a machine learning pipeline, where models come from, the AWS services for each stage, from AWS Glue and SageMaker AI to Amazon Bedrock, Amazon Quick, and Kiro, MLOps practices like tracked experiments, repeatable pipelines, monitoring, and retraining, and model metrics like accuracy, precision, recall, F1, and AUC next to business metrics like ROI. Includes three original practice questions.
Preview the topics
- AnyCompany’s data scientists have cleaned the claims data, and created features like the age of each policy. Which step of the pipeline comes next?
- The accuracy of AnyCompany’s fraud model has slowly dropped, because fraud patterns changed since its training. Which MLOps practice addresses this?
- AnyCompany’s fraud team says that missing a fraudulent claim costs far more than reviewing a legitimate one by mistake. Which metric matters most?