Skip to content
BytePatterns
Associate · MLA-C02Prep availableBeta

AWS Certified Machine Learning Engineer - Associate (MLA-C02) practice exam and study path

A free, timed 65-question MLA-C02 practice exam — 130 minutes, the same length as the real exam, scored by domain, with every option explained after you submit. Also 125 practice questions you can check one at a time, a page for every task in the exam guide, and a study path. No sign-up.

Start the free practice examGet the exam packPractise question by question →
Machine Learning Engineer - Associate exam pack:

$29once

For people who put ML into production: preparing data, training and tuning models, deploying and orchestrating them, and keeping them monitored and secure.

The exam

Exam code
MLA-C02
Questions
65 questions (50 scored)
Time
130 minutes
Passing score
720 of 1,000 (scaled)
Exam fee
$150 (USD)
Question types
multiple choice, multiple response
Beta format
85 questions, 170 minutes, $75

MLA-C02 has been in beta since September 29, 2026, in English only (booked as ME1-C02): 85 questions, 170 minutes, $75. The standard format above is the exam guide's 50 scored and 15 unscored questions; its time and fee are today's associate values, which AWS has not yet confirmed for MLA-C02 because general availability has not been announced. MLA-C01 ended in English on September 28, 2026 and stays available in Japanese, Korean and Simplified Chinese until MLA-C02 is generally available.

From AWS's certification page and exam guide, checked Oct 10, 2026. The exam is scored as a whole (compensatory): there is no pass mark per domain.

Domains and weights

  • Domain 1: Data Preparation for ML and AI

    28%

  • Domain 2: ML Model and Foundation Model (FM) Development

    24%

  • Domain 3: Deployment and Orchestration of ML and AI Workflows

    24%

  • Domain 4: Operating, Monitoring, and Securing ML and AI Solutions

    24%

On our roadmap this comes after Solutions Architect - Associate or AI Practitioner. See the whole roadmap.

Practice

125 original practice questions, every option explained with the AWS documentation page that proves it. Free, no sign-up; progress stays on this device.

Your results on this device →

More practice exams

The lessons, the sample questions on every task page and the first full practice exam are free and stay free. A pack adds the further timed MLA-C02 practice exams.

Machine Learning Engineer - Associate exam pack

$29once

  • Every paid MLA-C02 practice exam, timed, with every option explained
  • The next version of this exam included when it changes
  • Lifetime access — one payment, no subscription
  • 14-day refund, no conditions
  • 14-day refund, no conditions: ask within 14 days of buying and you get the whole payment back.
  • No pass guarantee. These are practice exams written by us from the exam guide and the AWS documentation; a score here does not predict your result on the real exam.

Study path

One module per exam domain. Lessons that already exist on BytePatterns are linked; the rest are being written.

  1. Module 1: Data Preparation for ML and AI

    1. 1Where ML data lives: S3, EFS, FSx, databases and choosing storage by cost, performance and complianceTask 1.1

      Partly covered by: S3: Consistency, Classes, Lifecycle, RDS vs DynamoDB

    2. 2File formats and ingestion: Parquet, ORC, JSON, CSV, Kinesis, Firehose and Managed Service for Apache FlinkTask 1.1

      Lesson coming

    3. 3Merging sources with AWS Glue and Spark on Amazon EMRTask 1.1, 1.2

      Lesson coming

    4. 4Vector stores for AI applications: OpenSearch Service, RDS with pgvector and Amazon S3 VectorsTask 1.1

      Partly covered by: Vector Databases, Approximate Neighbours

    5. 5SageMaker Feature Store: online and offline stores, ingestion and feature groupsTask 1.1, 1.2

      Lesson coming

    6. 6Transforming data with Glue, DataBrew, SageMaker Canvas data preparation and streaming Lambda functionsTask 1.2

      Partly covered by: Lambda & Event-Driven Design

    7. 7Feature engineering: scaling, standardization, encoding, binning, log transforms and feature splittingTask 1.2

      Lesson coming

    8. 8Text for models: tokenization, embeddings and domain-specific augmentationTask 1.2

      Partly covered by: Tokenization, BPE vs WordPiece, Embeddings, Cosine Similarity

    9. 9Preparing documents for RAG: chunking strategies and metadataTask 1.2

      Partly covered by: Chunking and Reranking, Retrieval-Augmented Generation

    10. 10Masking and redacting sensitive data; preparing data for fine-tuning, continued pre-training and distillationTask 1.2

      Partly covered by: Fine-Tuning vs Prompting

    11. 11Data quality with DataBrew and AWS Glue Data Quality; cleaning outliers, gaps and duplicatesTask 1.3

      Lesson coming

    12. 12Labeling, bias in data, class imbalance and validating training pairsTask 1.3

      Lesson coming

  2. Module 2: ML Model and Foundation Model (FM) Development

    1. 13Choosing an approach: AWS AI services, built-in algorithms, custom models or a foundation modelTask 2.1

      Partly covered by: What Is Machine Learning, What Is an LLM

    2. 14Selecting a foundation model in Amazon Bedrock and choosing between prompting, RAG and fine-tuningTask 2.1, 2.2

      Partly covered by: Fine-Tuning vs Prompting, Retrieval-Augmented Generation, Adapters and LoRA

    3. 15Tradeoffs: accuracy, interpretability, latency, training time and costTask 2.1

      Partly covered by: Training vs Inference

    4. 16Training on SageMaker AI: built-in algorithms, script mode and training jobsTask 2.2

      Lesson coming

    5. 17Automatic model tuning, early stopping and distributed trainingTask 2.2

      Lesson coming

    6. 18Overfitting, underfitting, regularization, catastrophic forgetting and ensemblesTask 2.2

      Lesson coming

    7. 19Tuning retrieval: embedding models, chunk size, hybrid search and rerankingTask 2.2

      Partly covered by: Chunking and Reranking, Embeddings

    8. 20Experiments and baselines: MLflow on SageMaker AI, shadow variants and explaining predictionsTask 2.3

      Lesson coming

    9. 21Metrics for classifiers and regressors: confusion matrix, precision, recall, F1, AUC, RMSETask 2.3

      Lesson coming

    10. 22Evaluating generative AI: BLEU, ROUGE, BERTScore, human review, LLM-as-a-judge and Amazon Bedrock evaluationsTask 2.3

      Partly covered by: Evaluating LLMs, LLM as a Judge

  3. Module 3: Deployment and Orchestration of ML and AI Workflows

    1. 23Inference options on SageMaker AI: real-time, serverless, asynchronous and batch transformTask 3.1

      Partly covered by: Training vs Inference

    2. 24Multi-model, multi-container and inference component endpointsTask 3.1

      Lesson coming

    3. 25Hosting foundation models: Bedrock on-demand, Provisioned Throughput, Custom Model Import and SageMaker AITask 3.1, 3.2

      Partly covered by: Quantization, The KV Cache

    4. 26Deploying agents: Amazon Bedrock AgentCore Runtime, Gateway, Memory and agent protocolsTask 3.1, 3.2

      Partly covered by: Agents and Tools, The Tool-Use Loop

    5. 27Amazon Bedrock Knowledge Bases and retrieval configurationTask 3.1, 3.2

      Partly covered by: Retrieval-Augmented Generation, Vector Databases

    6. 28Endpoints in a VPC, containers in ECR and deploying with the SageMaker Python SDKTask 3.2

      Partly covered by: VPC: Subnets, NAT & Firewalls, ECS vs EKS vs Fargate

    7. 29Auto scaling endpoints and GPU capacityTask 3.2

      Partly covered by: EC2, Auto Scaling & Load Balancers

    8. 30SageMaker Pipelines, Step Functions, MWAA and event-driven retrainingTask 3.3

      Partly covered by: SQS vs SNS vs EventBridge

    9. 31CI/CD for models: CodePipeline, CodeBuild, deployment guardrails and rollbackTask 3.3

      Lesson coming

    10. 32Versioning models, prompts and agents: Model Registry, MLflow, Prompt ManagementTask 3.3

      Lesson coming

  4. Module 4: Operating, Monitoring, and Securing ML and AI Solutions

    1. 33Monitoring models: data drift, prediction drift and A/B tests between variantsTask 4.1

      Lesson coming

    2. 34Observability for generative AI and agents: CloudWatch, AgentCore Observability and X-RayTask 4.1, 4.2

      Partly covered by: CloudWatch, Alarms & X-Ray

    3. 35Choosing instances and purchasing options for training and inferenceTask 4.2

      Partly covered by: AWS Cost Levers

    4. 36Foundation model costs: tokens, caching, batch inference and vector storageTask 4.2

      Partly covered by: Context Windows, Tokenization

    5. 37Least privilege for ML: execution roles, artifacts, Bedrock credentialsTask 4.3

      Partly covered by: Shared Responsibility & IAM

    6. 38Network isolation, encryption, auditing and scanning for ML systemsTask 4.3

      Partly covered by: VPC: Subnets, NAT & Firewalls

    7. 39Protecting sensitive data with Amazon Bedrock GuardrailsTask 4.3

      Partly covered by: Guardrails

Every task in the exam guide

The exam guide splits each domain into task statements. Each one has a page with a short summary, the lessons that teach it and open sample questions.

Domain 1: Data Preparation for ML and AI

Domain 2: ML Model and Foundation Model (FM) Development

Domain 3: Deployment and Orchestration of ML and AI Workflows

Domain 4: Operating, Monitoring, and Securing ML and AI Solutions

How these questions are made

  • Written by us from the exam guide's task statements and the AWS documentation — never from real exam content.
  • Every option carries its own explanation and a link to the AWS page that proves it.
  • A question enters a timed exam only after an independent check; until then it is a practice question.
  • Results are a plain percentage, per domain too — not an imitation of AWS's scaled score.

Questions about the MLA-C02 practice exam

Is the MLA-C02 practice exam free?
Yes. The full practice exam (65 questions, 130 minutes) is free with no sign-up, and so are the 125 practice questions, the task pages and the study path. Your attempts are saved in this browser, on this device.
How many questions does it have, and is it timed?
65 questions in 130 minutes: the same number of questions and the same time as the real MLA-C02 exam. When the time runs out, the exam is submitted as it stands, and unanswered questions count as wrong.
Are the answers explained?
After you submit, not during: like the real exam, the timed exam gives no feedback until the end. Then you get your score overall and per exam domain, and every question with each option explained and a link to the AWS documentation page behind it. In practice mode an answer is explained as soon as you check it.
What does the Machine Learning Engineer - Associate exam pack add?
3 more timed practice exams (Practice exam 1: 65 questions, 130 minutes; Practice exam 2: 65 questions, 130 minutes; Practice exam 3: 65 questions, 130 minutes), each scored and explained the same way, and the next version of this exam when it changes. One payment, lifetime access, no subscription. Everything above stays free. See the pack.
Can I get a refund?
Yes, with no conditions: ask within 14 days of buying and you get the whole payment back. The details are in the terms.
Is this an official AWS practice exam?
No. BytePatterns is not affiliated with, endorsed or sponsored by Amazon Web Services. Every question is written by us from the public exam guide and the AWS documentation, never from real exam content, and a score here does not predict your result on the real exam. AWS lists its own preparation resources on its certification page.