Exam Overview
Exam Details
- Code: MLA-C01 (MLA-C02 beta from Sep 2026)
- Level: Associate
- Duration: 130 minutes
- Questions: 65
- Passing Score: 720/1000
- Cost: ~$150 USD
- Validity: 3 years
👤 Who It's For
ML engineers, MLOps engineers, and developers who build and operationalize ML/GenAI solutions. AWS recommends ~1 year using Amazon SageMaker AI and related services, plus ~1 year in a related role (backend dev, DevOps, data engineer, or data scientist). A great next step after AI Practitioner (AIF-C01).
📊 Exam Domains
| Domain | Weight |
|---|---|
| 1. Data Preparation for Machine Learning | 28% |
| 2. ML Model Development | 26% |
| 3. Deployment and Orchestration of ML Workflows | 22% |
| 4. ML Solution Monitoring, Maintenance, and Security | 24% |
Domain weights reflect the MLA-C01 exam guide. MLA-C02 keeps the same domain structure but adds GenAI/agentic AI coverage. Confirm against the latest official guide.
📚 Key Topics
Data Preparation
- Feature engineering, transformation, and selection
- SageMaker Data Wrangler, Feature Store, Processing jobs
- Handling missing data, encoding, scaling, imbalance
- Data ingestion from S3, Glue, streaming sources
Model Development
- Algorithm selection; SageMaker built-in algorithms
- Training, hyperparameter tuning (AMT), evaluation metrics
- Overfitting/underfitting, cross-validation
- SageMaker JumpStart; foundation models via Bedrock (C02)
Deployment & Orchestration
- Real-time endpoints, batch transform, async & serverless inference
- SageMaker Pipelines; CI/CD for ML (MLOps)
- Model registry, versioning, A/B and shadow testing
- Infrastructure choices and cost optimization
Monitoring & Security
- SageMaker Model Monitor: data/model drift
- SageMaker Clarify: bias & explainability
- CloudWatch metrics/alarms; retraining triggers
- IAM, VPC, encryption for ML workloads
☁️ Key AWS Services to Know
| Category | Services |
|---|---|
| Core ML | Amazon SageMaker AI (Studio, Training, Endpoints, Pipelines, Model Monitor, Clarify, Feature Store, JumpStart) |
| Generative AI (C02) | Amazon Bedrock, Bedrock AgentCore, foundation models, RAG |
| Data | S3, Glue, Athena, Kinesis, Feature Store |
| Orchestration & MLOps | SageMaker Pipelines, Step Functions, CodePipeline, EventBridge |
| Monitoring & Security | CloudWatch, IAM, KMS, VPC, Model Monitor |
📋 Study Checklist
Progress0%
- Read the official MLA-C01 (or MLA-C02) exam guide
- Prepare data with SageMaker Data Wrangler & Feature Store
- Train and tune a model with SageMaker built-in algorithms + AMT
- Deploy real-time, batch, async, and serverless inference
- Build an MLOps pipeline with SageMaker Pipelines + model registry
- Configure Model Monitor for drift and Clarify for bias
- Explore generative AI via Amazon Bedrock (esp. for MLA-C02)
- Secure ML workloads: IAM, VPC, KMS encryption
- Complete the AWS Skill Builder ML Engineer learning plan
- Score 75%+ on practice exams before booking