AWS AI Certification Guide
Three tiers, one retired exam, and what it means for your team — 2026
AWS restructured its AI certification portfolio in 2026. The ML Specialty retired on March 31. Two new credentials replaced it — the ML Engineer Associate for MLOps work and the GenAI Developer Professional for Bedrock and generative AI applications. The AI Practitioner remains the foundational entry point. This guide covers all three, what changed, and how to navigate the new structure.
30-second executive takeaway
- The ML Specialty is gone. AWS retired the AWS Certified Machine Learning – Specialty on March 31, 2026. If your JDs still list it, update them. Candidates can no longer earn it. Map it to the ML Engineer Associate for MLOps skills or the GenAI Developer Professional for generative AI skills, depending on the role.
- AWS certs appear in 40% more job postings than competing vendor certs, tied with Google. They are the safest bet for engineers who want a credential that shows up on recruiter searches — whichever cloud platform your team runs on.
- The new GenAI Developer Professional covers Bedrock, RAG, and agentic patterns. It is the first AWS cert built specifically for the generative AI era, launched in 2026. Engineers building production LLM applications now have a dedicated AWS credential for that work.
CERT STACK
Three active certifications
AWS now has a clear three-tier structure for AI credentials. Each level targets a distinct role. They are not locked in sequence — an ML engineer with experience can go directly to the Associate without taking the Practitioner first.
AWS Certified AI Practitioner
The entry point. Covers AI/ML concepts, responsible AI, and AWS AI/ML service awareness. Built for people who work around AI — managers, analysts, product owners — not engineers building models. No hands-on prerequisites required.
AWS Certified Machine Learning Engineer – Associate
The mid-tier credential for engineers who build and operate ML systems on AWS. Requires hands-on experience with SageMaker, Bedrock, and the AWS ML service stack. Holders see approximately 20% salary premiums. Replaced the ML Specialty as the primary ML signal in job postings.
AWS Certified Generative AI Developer – Professional
New in 2026. The first AWS cert built specifically for the generative AI era — covers building production GenAI apps on Bedrock, RAG patterns, knowledge bases, agentic architectures, and responsible deployment. This is the credential for engineers shipping LLM-backed products, not managing traditional ML pipelines.
RETIRED
What happened to the ML Specialty
The AWS Certified Machine Learning – Specialty was retired on March 31, 2026. For several years it was the primary signal for ML engineering depth on AWS — a demanding exam that required production SageMaker experience and covered the full ML lifecycle from data prep through model deployment.
AWS retired it because the ML landscape fragmented. The ML Specialty tried to cover traditional ML and emerging generative AI work in a single exam, which made it an awkward fit for both. The replacement structure is more precise: ML Engineer Associate for engineers running SageMaker Pipelines and model monitoring, GenAI Developer Professional for engineers building Bedrock applications.
Existing ML Specialty holders keep their certification until it expires — AWS does not revoke it retroactively. But no new candidates can earn it. If you have it on your resume, it remains valid and recognized. If you are hiring for it, you need a different signal now.
Mapping the retirement
STUDY PATH
How to prepare for AWS AI certifications
AWS certifications reward hands-on experience over documentation memorization. The study path below applies to the ML Engineer Associate and GenAI Developer Professional — the two exams that require real AWS experience to pass.
Apply for the AWS AI & ML Scholars program (if still open)
Applications are open through June 24, 2026. The program provides free training toward the AI Practitioner certification. Even if you are targeting the Associate or Professional tier, the free training covers foundational AWS ML service concepts that appear across all three exams.
Build hands-on with SageMaker and Bedrock in an AWS sandbox
For the ML Engineer Associate, you need real SageMaker experience — pipeline creation, training jobs, endpoint deployment, and model monitoring. For the GenAI Developer Professional, build at least one production-grade application on Bedrock using knowledge bases and agents. Sandbox accounts are available through AWS Free Tier and AWS Skill Builder.
Study the ML Engineer Associate exam guide — focus on MLOps
The ML Engineer Associate exam covers the full MLOps lifecycle: data ingestion and preparation, SageMaker Pipelines, training optimization, model evaluation, deployment strategies (blue/green, A/B), and production monitoring with SageMaker Model Monitor. The exam guide is public on aws.amazon.com/certification — read it before setting your study schedule.
For the GenAI Developer: build a RAG application on Bedrock
The GenAI Developer Professional tests real architectural decisions about RAG implementation — chunking strategies, embedding models, knowledge base configuration, retrieval tuning, and response generation. Build a working RAG application with Bedrock Knowledge Bases and Bedrock Agents. The exam scenario questions require the kind of judgment you only get from having debugged retrieval quality issues in a real application.
Take AWS Skill Builder practice exams before your test date
AWS Skill Builder includes official practice exams for all three AI certifications. The practice questions use the same scenario-based format as the real exam. A Skill Builder subscription includes full access to practice exams across all AWS certifications — worthwhile if you are preparing for multiple exams or want to build confidence before your test date.
MARKET SIGNAL
AWS certs in the job market
AWS certifications appear in job postings at roughly the same rate as Google Cloud certifications — both show up in approximately 40% more postings than other vendor AI certs. That frequency reflects market reality: AWS is the dominant cloud platform globally, and a meaningful share of ML and GenAI engineering work runs on it.
- The ML Specialty retirement created a gap in recruiter search filters. Many ATS systems and Boolean search strings were configured to find "AWS Machine Learning Specialty." Those searches now return fewer results. If you are sourcing ML engineering talent, update your search terms to "ML Engineer Associate" or remove the cert requirement entirely and screen for SageMaker and Bedrock experience directly.
- The ML Engineer Associate carries approximately 20% salary premiums for holders compared to peers without AWS ML credentials, based on compensation data across cloud-native engineering roles. The effect is strongest at AWS-heavy organizations and consulting firms where the cert is treated as a hiring filter rather than a differentiator.
- The GenAI Developer Professional is too new for reliable salary data, but professional-level AWS certs historically command stronger premiums than associate-level credentials. As Bedrock adoption grows, the GenAI Developer Professional is likely to become the dominant signal for generative AI engineering work on AWS.
RELATED GUIDES
More certification guides
All AI Certifications
25+ credentials from 14 providers compared by cost, format, and hiring signal.
Google AI Certification
Google Cloud ML Engineer, Gemini certification, and the full Google credential stack.
Anthropic Certification
CCA-F exam and Anthropic Academy — what they test and what they signal to employers.
Frequently Asked Questions
What AWS AI certifications are available in 2026?
What happened to the AWS ML Specialty certification?
How much do AWS AI certifications cost?
Which AWS AI cert should I get first?
Does the AWS ML Engineer cert increase salary?
How does AWS compare to Google and Microsoft AI certifications?
See how AWS stacks up against the rest
The full guide covers 25+ credentials from 14 providers, compared by cost, format, and what they actually signal to employers.