AWS · MLA-C02 · BETA
Infrastructure, containers, and VPC endpoints
MLA-C02 · Lesson 16 of 23 · 2:07:06
YouTube loads only after you press play. This chapter opens at its timestamp in the full course. Watch on YouTube ↗
In this lesson
Lesson 16 of the free MLA-C02 beta series (AWS Certified Machine Learning Engineer – Associate, domain 3, deployment and orchestration). In this lesson: infrastructure as code with CloudFormation and the CDK and stacks that work together, containers for ML (use, extend, or build), images in Amazon ECR, and SageMaker AI endpoints in a VPC with network isolation. With three quick checks and a recap, and AnyCompany Mobility, a car sharing service, as the running example.
Preview the topics
- The platform team's network stack creates the private subnets and the security group. The ML team's endpoint stack must use them in every environment, without copying IDs by hand. What should the ML engineer do?
- A team pushes every build with the tag latest, and nobody can tell which image a model was tested with. The release process needs tags that always point to the same image. What should the ML engineer turn on?
- A SageMaker AI endpoint uses private subnets with no route to the internet. It fails to start, because the model artifacts in Amazon S3 cannot be loaded. What should the ML engineer add?