Most candidates need 8 to 10 weeks of part-time study to pass the AWS Certified Machine Learning Engineer Associate (MLA-C01) exam. Engineers already shipping models on Amazon SageMaker often need 4 to 6 weeks, while those new to SageMaker should budget 12 to 14 weeks. One date matters more than any of these: MLA-C01 retires in English on September 28, 2026.
Last updated August 2026.
This certification is mid-changeover, so your study timeline decides which exam you sit. Registration for the updated exam, MLA-C02, opens in English on September 1, 2026, and the last day to take MLA-C01 in English is September 28, 2026. For all exam languages, the standard version of MLA-C02 arrives in early 2027.
Do the arithmetic before you start. From early August, a typical 8 to 10 week plan finishes right at or just past the MLA-C01 cutoff. That leaves two honest options: compress to hit the deadline on a settled exam, or aim past it and prepare for MLA-C02, which brings generative AI, agentic AI, and foundation model workloads into scope alongside traditional ML engineering.
The rule of thumb: if you are starting from real SageMaker experience, MLA-C01 is comfortably reachable. If you are starting from scratch, do not rush a compressed plan to beat the cutoff. Study for MLA-C02 and use the extra weeks properly.
Changeover dates from the AWS Training and Certification announcement for MLA-C02. Everything below describes MLA-C01.
These assume 7 to 8 hours a week, so the typical case lands near 60 to 80 hours total. That is longer than most associate exams, and the exam guide explains why: its target candidate has at least one year using Amazon SageMaker and other AWS services for ML engineering, plus at least one year in a related role such as backend developer, DevOps engineer, data engineer, or data scientist. It expects two kinds of experience, not one.
Recommended experience is from the official MLA-C01 exam guide. It describes the target candidate, it is not an entry requirement.
Most AWS exams have a lightest domain you can under-study and still pass. MLA-C01 does not: its four domains sit in a narrow band from 22% to 28%, so the gap between biggest and smallest is about three scored questions.
| Domain | Weight | Of a 70-hour plan |
|---|---|---|
| 1. Data Preparation for Machine Learning | 28% | About 20 hours |
| 2. ML Model Development | 26% | About 18 hours |
| 4. ML Solution Monitoring, Maintenance, and Security | 24% | About 17 hours |
| 3. Deployment and Orchestration of ML Workflows | 22% | About 15 hours |
Domain weights from the official MLA-C01 exam guide. Full breakdown in the domains and weighting post.
That flatness is the scheduling problem. A plan that front-loads modeling and leaves monitoring and security for the final weekend is betting on a domain worth about 12 scored questions, and compensatory scoring will not rescue a blank one.
Compress or stretch: shipping models on SageMaker already? Fold Weeks 1 to 6 into three weeks of targeted review and keep Weeks 7 to 9 intact. New to SageMaker? Add a week to each of the first three blocks and expect the hands-on week to be the most valuable one.
Most candidates need 8 to 10 weeks of part-time study, roughly 60 to 80 hours. Engineers already shipping models on SageMaker often need 4 to 6 weeks, and those new to SageMaker should budget 12 to 14 weeks. MLA-C01 has no light domain to skim, which is what stretches the timeline.
If you can be ready by September 28, 2026, sitting MLA-C01 means studying a settled exam with a published guide. If your realistic timeline runs past that date, plan for MLA-C02 and study its guide instead, since it adds generative AI, agentic AI, and foundation model workloads to the scope.
Plan for 60 to 80 hours if you have some SageMaker exposure, less if you work with it daily and more if it is new to you. Across 8 to 10 weeks that is about 7 to 8 hours a week.
No, it is a recommendation rather than a requirement. The exam guide describes its target candidate as having at least one year using Amazon SageMaker and other AWS services for ML engineering, plus at least one year in a related role. Less experience mainly means a longer study runway.
The guide lists the task statements and in-scope services behind all four domains. Check the version that matches your test date before you build a plan around it.
Open the AWS exam guide →