OtanoAcademy
← All study guides
AWS Certified Machine Learning Engineer · Associate badge
MLA-C02 · Free study guideBeta

AWS Certified Machine Learning Engineer · Associate

For ML engineers: you prepare data, build or customize models (including foundation models), deploy them, and keep them monitored, efficient and secure. MLA-C02 is in beta in English.

Questions
85 (beta)
Time
170 minutes
Passing score
720 / 1000
Level
Associate

What the exam covers

Domains and key topics

Domain 128% of the exam

Data Preparation for ML and AI

  • Collecting and storing training data: S3, data formats, feature stores
  • Feature engineering and pre-processing with SageMaker Data Wrangler and Glue
  • Data quality and bias detection with SageMaker Clarify
  • Labeling data with SageMaker Ground Truth
Official domain details ↗
Domain 224% of the exam

ML Model and Foundation Model (FM) Development

  • Choosing an approach: built-in algorithms, custom models, foundation models
  • Training, tuning and fine-tuning: hyperparameters, SageMaker training jobs
  • Customizing foundation models with Amazon Bedrock and SageMaker JumpStart
  • Evaluating models: metrics, overfitting, model and FM evaluation
Official domain details ↗
Domain 324% of the exam

Deployment and Orchestration of ML and AI Workflows

  • Endpoint types: real-time, serverless, asynchronous, batch transform
  • Provisioning compute for training and inference, including accelerators
  • MLOps pipelines: SageMaker Pipelines, Model Registry, CI/CD
Official domain details ↗
Domain 424% of the exam

Operating, Monitoring, and Securing ML and AI Solutions

  • Monitoring drift and quality with SageMaker Model Monitor
  • Cost and performance: right-sizing, Spot training, auto scaling endpoints
  • Securing endpoints and data: IAM, VPC, encryption, network isolation
Official domain details ↗

Know these well

Key AWS services

Amazon SageMaker AI

Build, train, tune and deploy ML models.

SageMaker Pipelines & Model Registry

Repeatable MLOps workflows and model versions.

SageMaker Clarify & Model Monitor

Bias, explainability and drift.

SageMaker Data Wrangler & Feature Store

Prepare and reuse features.

SageMaker JumpStart

Pretrained and foundation models to start from.

Amazon Bedrock

Managed foundation models and customization.

AWS Glue

Data preparation at scale.

Amazon S3

Training data and model artifacts.

Amazon CloudWatch

Metrics and alarms for endpoints.

AWS KMS & Amazon VPC

Encryption and private networking for ML.

How to prepare

A simple study plan

  1. 1Read the official MLA-C02 exam guide and its comparison with MLA-C01.
  2. 2Run one model through SageMaker from data preparation to a deployed endpoint.
  3. 3Learn the endpoint types and when each one fits.
  4. 4Study foundation model customization: when to fine-tune and when to use retrieval.
  5. 5Take a timed practice exam and review every explanation, right and wrong.

Always check the official MLA-C02 exam guide ↗ for the current version of the exam.

Preparing for more than one exam?

Every AWS certification has a free study guide here, from Foundational to Specialty.

All study guides →