AWS Certified Machine Learning Engineer - Associate (MLA-C02) Study Guide [280 Minute Read]

A First-Principles Approach to Machine Learning Engineering on AWS

Welcome to the AWS Certified Machine Learning Engineer - Associate (MLA-C02) Study Guide. This guide moves beyond surface-level memorization. It is designed to build a robust mental model of how machine learning systems are built, deployed, and maintained on AWS—understanding the why behind every architectural decision, service selection, and operational trade-off.

Each topic is aligned with the official AWS MLA-C02 exam guide, targeting the specific cognitive skills required for success. Expect scenario-based questions that test your ability to choose the right AWS service for a given ML workflow stage, troubleshoot pipeline failures, and make cost-performance trade-offs—roughly 60% application, 30% analysis, and 10% recall.

Exam Details (MLA-C02): 65 questions (50 scored, 15 unscored) — multiple choice and multiple response only (multiple response = two or more correct answers out of five or more options) | Passing score: 720/1000 | AWS has not yet published the duration of the general-availability exam; the beta ran 85 questions in 170 minutes.

⚠️ Exam version notice: AWS is replacing MLA-C01 with MLA-C02. English delivery of MLA-C01 ends on 28 September 2026; the MLA-C02 beta runs from 29 September 2026 and MLA-C02 reaches general availability in all languages on 14 January 2027. MLA-C02 keeps the four domains of MLA-C01 and adds generative AI, agentic AI and foundation-model workloads (Amazon Bedrock, retrieval-augmented generation, agents, FM evaluation and guardrails). This guide and its practice questions are being updated for MLA-C02 — the new generative-AI sections are rolling out over the coming weeks, and the traditional ML engineering content below remains in scope.

Prerequisites: At least 1 year of experience using Amazon SageMaker AI, Amazon Bedrock, and other AWS services for ML engineering, with hands-on experience of both traditional ML and generative AI. Familiarity with common ML algorithms, data engineering fundamentals, CI/CD pipelines, and software engineering best practices. Background in a role such as backend developer, DevOps engineer, data engineer, or data scientist is assumed.

Exam Domain Weights

Domain 1 (Data Preparation) carries the heaviest weight at 28%, reflecting the reality that most ML engineering effort goes into getting data right before models ever train. Combined with Domain 4's 24% on monitoring and security, over half the exam tests your ability to handle what happens around the model—not just the model itself. Prioritize these operational areas alongside model development.

---