AWS Certified Machine Learning Engineer – Associate — free practice exam
82 original practice questions aligned to official exam domains. Run a timed, domain-weighted quiz in your browser — no login required.
Exam domains
| Domain | Weight |
|---|---|
| Domain 1: Data Preparation for Machine Learning | 28% |
| Domain 2: ML Model Development | 26% |
| Domain 3: Deployment and Orchestration of ML Workflows | 24% |
| Domain 4: ML Solution Monitoring, Maintenance, and Security | 22% |
Sample practice questions
- What is the MOST appropriate AWS solution to use built-in algorithms for common ML tasks?
- What is the MOST appropriate AWS solution to deploy real-time inference behind an endpoint?
- What is the MOST appropriate AWS solution to deploy multiple models behind one endpoint?
- What is the MOST appropriate AWS solution to detect schema and statistics on S3 data?
- What is the MOST appropriate AWS solution to detect data drift in production features?
- What is the MOST appropriate AWS solution to encrypt training volumes and model artifacts?
- What is the MOST appropriate AWS solution to explain individual predictions to stakeholders?
- What is the MOST appropriate AWS solution to handle class imbalance in training data?
- What is the MOST appropriate AWS solution to label images for object detection?
- What is the MOST appropriate AWS solution to orchestrate ML CI/CD steps?
- What is the MOST appropriate AWS solution to run ETL on large datasets in a data lake?
- What is the MOST appropriate AWS solution to run asynchronous inference for large payloads?
- What is the MOST appropriate AWS solution to track experiments and parameters?
- What is the MOST appropriate AWS solution to train models with managed infrastructure and spot savings?
- What is the MOST appropriate AWS solution to tune hyperparameters automatically?
- What is the MOST appropriate AWS solution to visually prepare tabular data for ML without code?
- An audit found gaps: the organization must schedule batch scoring nightly. Which option addresses this? schedule batch…
- An audit found gaps: the organization must audit who invoked endpoints. Which option addresses this? audit who invoked…
- A company needs to capture inference payloads for audit (where allowed). Which option meets the requirement with the…
- A company needs to distributed training across multiple GPUs. Which option meets the requirement with the LEAST…
- A company needs to track experiments and parameters. Which option meets the requirement with the LEAST implementation…
- For cost and security, the team must use infrastructure as code for SageMaker resources. What is the recommended…
- For cost and security, the team must detect data drift in production features. What is the recommended approach?…
- For cost and security, the team must scan containers for CVEs before deploy. What is the recommended approach? scan…
Browse all 82 indexed questions for this exam.