AWS machine learning certification: which one to take

Which AWS AI and ML certifications exist now?

AWS now has three AI and ML certifications for technical people. The AWS Certified AI Practitioner (AIF-C01) is the foundational exam. The AWS Certified Machine Learning Engineer - Associate covers building and running ML in production. The AWS Certified Generative AI Developer - Professional (AIP-C01) covers production applications built on foundation models. AWS also lists an AWS Certified AI Business Strategist. That one tests business judgment on AI investment, governance and adoption, and AWS says it does not test knowledge of AWS AI services.

The AWS Certified Machine Learning - Specialty has been retired. The last day to sit it was March 31, 2026. If you already hold it, it stays active for three years from the date you earned it. AWS has pointed former Specialty candidates to the Practitioner, the ML Engineer Associate, the Data Engineer Associate and the Generative AI Developer Professional.

What is changing on the ML Engineer Associate exam?

The exam is moving from MLA-C01 to MLA-C02. The last English sitting of MLA-C01 was September 28, 2026, and the MLA-C02 beta started on September 29. The beta is English only and costs $75 for 85 questions in 170 minutes. MLA-C01 cost $150 for 65 questions in 130 minutes. It stays open in Japanese, Korean and Simplified Chinese until MLA-C02 reaches general availability, which AWS expects in early 2027 but has not yet dated. At that point MLA-C02 opens in all languages.

The domains are the same. What AWS added is generative AI on Amazon Bedrock, retrieval-augmented generation, agentic workflows, choosing and fine-tuning foundation models, and responsible AI. AWS now recommends a year of experience with Amazon SageMaker AI and Amazon Bedrock, plus a year in a related role such as backend developer, DevOps, data engineer or data scientist. A certificate earned on MLA-C01 stays valid until its original expiry date.

Which one fits you?

If you work near AI but don't build it, the AI Practitioner suits you. That covers product, sales, IT and business roles, and AWS aims it at people with up to six months of exposure. If you train, deploy and monitor models on AWS, or you are a data scientist moving into engineering, take the ML Engineer Associate. If you already ship LLM applications, take the Generative AI Developer Professional. AWS describes that candidate as having two or more years of production work plus one year of hands-on generative AI.

A certification shows you know the AWS services. It does not stand in for a portfolio of models and pipelines you have actually shipped, and it guarantees no job. It helps most when an employer or consulting partner runs on AWS and screens for AWS credentials.

Pay

BLS software developers, 25th percentile to median

$108,500–140,200

BLS software developers, median to 75th percentile

$140,200–177,400

AutomationRoles.ai ML engineer job ads, 25th percentile to median

$184,000–227,400

AutomationRoles.ai ML engineer job ads, median to 75th percentile

$227,400–267,900

USD a year, before tax. US figures. BLS rows use OEWS May 2025 for software developers, moved to 2026 with the Employment Cost Index; it is the nearest series and covers the whole occupation, not ML specialists alone. Job ad rows are midpoints of US dollar ranges in 139 live AutomationRoles.ai ads from 56 employers on 5 November 2026; they lean toward large AI and technology employers and leave out stock and bonus. Refer to the Machine Learning Engineers salary guide for figures from live adverts.

What you need

AI Practitioner (AIF-C01)Foundational. 65 questions, 90 minutes, $100 USD. Aimed at people with up to six months of exposure to AI and ML on AWS who use these tools without necessarily building them.
ML Engineer Associate, MLA-C02 beta85 questions, 170 minutes, $75 USD at beta pricing, English only. Aimed at people with at least one year using SageMaker AI, Bedrock and other AWS ML services.
ML Engineer Associate, MLA-C0165 questions, 130 minutes, $150 USD. Closed in English since September 28, 2026. Still offered in Japanese, Korean and Simplified Chinese until MLA-C02 general availability, which AWS expects in early 2027.
Generative AI Developer Professional (AIP-C01)75 questions, 180 minutes, $300 USD. Aimed at people with two or more years building production applications and one year of hands-on generative AI work.
No prerequisitesAWS does not require you to hold another certification before sitting any of these exams. All of them run at a Pearson VUE test center or as an online proctored exam.
Validity and recertificationEach certification lasts three years. Passing the ML Engineer Associate or the Generative AI Developer Professional also renews an AI Practitioner certification, and passing the Professional renews the ML Engineer Associate.

How to get there

Year 0-1Build hands-on AWS ML experienceTrain, deploy and monitor models with SageMaker AI and build on Bedrock. AWS's recommended experience for the associate exam is about a year of this work.
Months 0-6Optional: AI PractitionerTake this only if you are new to AI on AWS or work in a non-building role. Experienced engineers can skip it.
Year 1-2ML Engineer AssociateTake the MLA-C02 beta in English for $75, or wait for the standard release in all languages, which AWS expects in early 2027.
Year 2-3Generative AI Developer ProfessionalThis one fits once you have about two years of production work and a year of shipping generative AI. Passing it also renews your associate certification.
Every 3 yearsRecertifyEach certification expires after three years. Renew it by passing the latest version of the exam or a higher-level exam that AWS accepts for renewal.

Open now on AutomationRoles.ai

Senior Machine Learning Engineer, AI Safety EvaluationsNVIDIA · Santa Clara, California

Machine Learning Engineer 4 - Manager ICCapital One · New York, New York

Machine Learning Engineer, Core ExperimentationOpenAI · Seattle, Washington

Machine Learning Engineer 4Capital One · McLean, Virginia

Machine Learning Engineer, FAIAmazon · Sunnyvale, California

See all →

Questions people ask

Can I still take the AWS Certified Machine Learning - Specialty?

No. AWS retired it, and the last day to take the exam was March 31, 2026. If you already hold it, it stays active for three years from the date you earned it.

Should I take the MLA-C02 beta or wait for the standard exam?

The beta costs $75, runs only in English and has 85 questions in 170 minutes. Some of those questions are unscored items AWS uses for statistics. AWS says the standard MLA-C02 will be available in all supported languages in early 2027 and has not given an exact date. AWS has not said on the pages reviewed here what the standard version will cost.

Is my MLA-C01 certification still valid?

Yes. A certificate earned on MLA-C01 stays active through its original three-year validity period, no matter when the exam is retired.

Which AWS certification suits a data scientist?

AWS lists data scientists moving into ML engineering among the target roles for the ML Engineer Associate. AWS also recommends a year of related experience and a year using AWS ML services before you sit it.

Will an AWS certification get me hired as an ML engineer?

No certification guarantees a job. It shows an employer that you know the AWS ML services. It counts most at companies and consulting partners that run on AWS, and it works best alongside projects you have actually shipped.

Which AWS machine learning certification should you take?