
This course emphasizes hands-on learning: each module contains labs, guided demos, interactive exercises, and practice exams to reinforce applied skills on AWS.
- Domain 1 — Data processing for ML: ingesting, storing, transforming, and validating data with Amazon S3, Amazon Kinesis, AWS Glue, and SageMaker Data Wrangler.
- Domain 2 — ML development: choosing algorithms, automated training, hyperparameter optimization, and working with foundation models via SageMaker built-in algorithms and SageMaker JumpStart.

- Domain 3 — Deployment and orchestration: implementing real-time, batch, serverless, and asynchronous inference patterns and integrating them into production workflows.

- Domain 4 — Monitoring, maintenance, and security: model and data monitoring (for example, using SageMaker Model Monitor to detect data and model drift), bias detection and explainability (SageMaker Clarify), plus security and governance for ML systems.

Key learning outcomes from this course:
- Build reliable data pipelines for ML on AWS.
- Train, tune, and validate models using SageMaker tooling.
- Deploy models using real-time, batch, and serverless patterns.
- Monitor model health and data drift; apply explainability and bias detection.
- Apply security best practices to protect ML workflows and artifacts.
- AWS Documentation — Amazon SageMaker
- SageMaker Model Monitor
- SageMaker Clarify
- AWS Certified Machine Learning — Associate Exam Guide (review for exam specifics and objectives)
Hands-on practice is essential. The exam tests practical knowledge of AWS services and design patterns — complete the labs and practice exams included in the course to prepare effectively.