Machine Learning Engineer
Prepare data, develop models, deploy inference and monitor the behavior of machine learning workloads.
Who this path is for
Know Python, model evaluation and basic AWS services.
Study each topic, complete the applied exercise, and keep evidence of what you learned. These are original study milestones, not a claim to reproduce the full exam blueprint. Use the linked AWS exam guide to check all in-scope domains before booking.
Your learning roadmap
01Prepare ML data
3 focus areas · applied exercise
Prepare ML data
3 focus areas · applied exercise- Cleaning and features
- Training/validation splits
- Data leakage
Design a feature pipeline for an order-demand prediction task.
Show how the split prevents future information leaking into training.
02Develop and evaluate
3 focus areas · applied exercise
Develop and evaluate
3 focus areas · applied exercise- SageMaker AI
- Training jobs
- Model selection
Compare candidate models using an explicit baseline.
Explain the chosen metric and the cost of a false prediction.
03Deploy inference
3 focus areas · applied exercise
Deploy inference
3 focus areas · applied exercise- Batch and online inference
- Model packaging
- Release strategy
Diagram an endpoint release with rollback and access controls.
Specify latency and availability requirements.
04Monitor and maintain
3 focus areas · applied exercise
Monitor and maintain
3 focus areas · applied exercise- Model drift
- Pipelines
- Security and cost
Define a drift investigation and retraining decision process.
Document an alert threshold, owner and recovery action.
Put the learning into practice
Related labs cover selected skills. Completing them does not establish exam readiness.