AWS - Machine Learning Engineer Associate Study Guide

3: Deployment and Orchestration of ML Workflows

This chapter takes trained model artifacts from the SageMaker Model Registry and turns them into production inference systems on AWS. It covers deployment target selection across SageMaker AI, EKS, ECS, and Lambda; the four SageMaker inference options; auto scaling and networking for endpoints; container hosting patterns with ECR; and the CI/CD stack that drives repeatable ML delivery through SageMaker Pipelines, CodePipeline, EventBridge, and infrastructure as code.