MLA-C01 vs. MLA-C02 Beta: Which AWS Machine Learning Engineer Exam Should You Take?

AWS has opened registration for the updated AWS Certified Machine Learning Engineer - Associate exam, MLA-C02. For English-speaking candidates, that creates an immediate decision: take MLA-C01 before it retires, or wait for the new beta exam.

There is no single answer. The better option depends on how far along you are, what skills you want the certification to reflect, and how comfortable you are with a beta exam.

The timeline matters

MLA-C01 remains available in English until September 28, 2026. The MLA-C02 beta begins on September 29. The standard MLA-C02 exam is scheduled for general availability on January 14, 2027.

AWS will keep MLA-C01 available in Japanese, Korean, and Simplified Chinese during the beta period. This article focuses on English-speaking candidates because the transition is much more immediate.

What is changing in MLA-C02?

MLA-C01 focuses on machine learning engineering and MLOps. MLA-C02 retains those foundations but expands the scope to match current production work with generative AI.

The new version adds coverage of:

●Generative AI implementation and deployment

●Foundation models and large language models

●Retrieval-augmented generation architectures

●Agentic AI workflows

●Amazon Bedrock

●Responsible AI practices

The exam still targets people who build, deploy, maintain, and monitor machine learning workloads on AWS. It is not becoming a purely generative AI certification. Traditional machine learning engineering remains part of the credential.

Take MLA-C01 if you are already prepared

MLA-C01 is usually the sensible choice if you have already invested serious time in its objectives, training materials, labs, and practice exams.

Waiting for MLA-C02 may force you to pause a study plan that is close to completion. If you can pass MLA-C01 before September 28, you still earn the AWS Certified Machine Learning Engineer - Associate credential. AWS states that the credential remains active through its original expiration date.

This route is especially practical for candidates who:

●Have already completed an MLA-C01 course

●Have consistently passed MLA-C01 practice exams

●Need the certification for a current job, project, or employer deadline

●Prefer a standard exam with established preparation resources

Taking MLA-C01 does not mean your knowledge is immediately outdated. It means you have validated the version of the skill set that was current when you earned the certification. You can then add GenAI and Amazon Bedrock experience through hands-on work, AWS Skill Builder, or later recertification.

Choose MLA-C02 Beta if GenAI is central to your work

The MLA-C02 beta is more attractive if your current or intended role involves LLMOps, Amazon Bedrock, RAG, AI agents, or production generative AI applications.

The updated exam better reflects how ML engineering roles are changing. A professional who already works with foundation models, evaluation, fine-tuning, retrieval pipelines, and AI governance may find that the new objectives fit their day-to-day responsibilities more closely.

The beta is also an option for candidates who are not yet deeply committed to MLA-C01 study materials. If you are still early in preparation, switching now may be more efficient than preparing for an exam that will soon retire in English.

Understand the trade-off of a beta exam

Beta exams have advantages, but they involve uncertainty.

MLA-C02 beta is available only in English, costs $75 USD, lasts 170 minutes, and includes 85 questions. Because it is a beta exam, preparation resources and practice tests will be less mature than those available for MLA-C01.

Candidates should also expect that the final scoring process may take longer than a standard exam. The beta is best for people who are comfortable studying from the official exam guide, AWS documentation, and hands-on labs instead of depending on a large ecosystem of third-party mock exams.

A practical decision rule

Choose MLA-C01 if you are close to ready and can take the exam before September 28.

Choose MLA-C02 Beta if you are early in preparation or want your certification study to focus on generative AI, LLMs, Amazon Bedrock, RAG, and agentic workflows.

In either case, the most useful preparation is not limited to passing an exam. Build and operate ML workloads, work through AWS labs, and be able to explain your design choices. The version number may change, but practical engineering judgment remains useful across both exams.