AWS Certified Generative AI Developer Professional: Exam Cost, Format and Who It Fits

The AWS Certified Generative AI Developer Professional is AWS’s top-level exam for building generative AI applications. It costs $300 USD, has 75 questions and runs 180 minutes. It’s for developers with about 2 years of production experience on AWS or open-source tools and at least a year of hands-on generative AI work. It isn’t a first certification.

What Is the AWS Certified Generative AI Developer Professional?

It’s a professional-level exam from Amazon Web Services. Professional is the highest level among the AWS AI exams we cover, above the foundational AI Practitioner and the associate-level Machine Learning Engineer exam.

The target candidate is a developer, not a manager or a student. AWS describes someone who has built production-grade applications, has general AI/ML or data engineering experience and has spent a year implementing generative AI solutions. In plain terms, you’ve already shipped something with a foundation model, and the exam checks that you did it well.

Who it fits:

• Software engineers who build applications on top of foundation models.
• Machine learning engineers who’ve moved into generative AI work.
• Cloud and data engineers who now run generative AI workloads.
• Consultants who deliver generative AI projects on AWS.

If you’re still learning the basics, the AWS Certified AI Practitioner is a better first step.

How Much Does the Exam Cost?

AWS lists the exam at $300 USD. That’s three times the AI Practitioner’s $100 and double the $150 for the current Machine Learning Engineer Associate exam. It matches the final price of the retired Machine Learning Specialty.

Your real cost goes beyond the fee. If you need to build the experience AWS recommends, that takes months of work, and if you pay for practice exams or a prep course, add that too. A failed attempt means paying the fee again, so it’s worth preparing well before you book.

Check the price on AWS’s certification page when you book, since taxes and local pricing can change what you pay.

What Is the Exam Format?

AWS publishes this format:

• Questions: 75, multiple choice or multiple response.
• Time: 180 minutes.
• Delivery: Pearson VUE testing center or online proctored.

At 180 minutes, it’s tied for the longest of the AWS AI exams we cover. Three hours is a long time to stay sharp, so plan your pacing and your breaks around it. With 75 questions in 180 minutes, you have a bit over two minutes for each one, and professional-level questions tend to be scenarios with several plausible answers.

Multiple-response questions ask for more than one correct choice. Read the instructions on each one so you don’t pick too few.

Before you plan your study time, open AWS’s certification page and read what it says each part of the exam tests, so you know where to spend your hours.

What Experience Does AWS Recommend?

AWS says the target candidate has:

• 2 or more years of experience building production-grade applications on AWS or with open-source technologies.
• General AI/ML or data engineering experience.
• 1 year of hands-on experience implementing generative AI solutions.

These are recommendations, not booking requirements. They still tell you something. This exam assumes you’ve handled the unglamorous parts of a real application: authentication, cost, failures, monitoring and change over time. If you haven’t, a professional exam is likely to feel like a wall.

What Does the Exam Test?

AWS frames it around building generative AI solutions in production. The experience it recommends gives you a good picture: working applications that use foundation models, built on AWS or open-source tools, by someone who also understands the data and machine learning around them.

For the detailed list of domains and weights, use the exam guide on AWS’s certification page. That’s where AWS states them, and where the weights are kept current.

To get a sense of the neighbors on the ladder, note what AWS says about the associate exam. Its updated version, MLA-C02, adds generative AI implementation, including Amazon Bedrock and retrieval-augmented generation architectures, plus agentic AI and foundation models. A professional exam on generative AI development will ask you to go further than that: to choose, build and run these systems, not just recognize them. Our Machine Learning Engineer Associate page covers the version change.

How Do You Prepare?

Start from what AWS says the exam tests, then fill the experience gaps you find.

1. Read AWS’s description and mark every task you haven’t done. On a professional exam, an honest gap list is your most useful tool.
2. Build something real. Pick a small generative AI application and put it on AWS. Work through data handling, calling a model, retrieval, security and monitoring.
3. Break it on purpose. Test what happens when a model call fails, when costs spike or when an answer goes wrong. Scenario questions reward people who’ve seen failures.
4. Work with your team’s real systems if you can. Production experience is the thing AWS recommends and the thing practice tests can’t supply.
5. Practice with scenario questions. They show you how long the questions run and how AWS words trade-offs.
6. Rehearse the three hours. Do at least one timed, full-length practice session so test day isn’t your first long sitting.

Give yourself more time than you’d give a foundational exam. For most people, the work is in the experience, not in memorizing.

How Does It Compare With the Other AWS AI Exams?

Here is the ladder, using the figures AWS publishes.

• AWS Certified AI Practitioner: foundational, $100, 65 questions, 90 minutes. No prerequisite. See the AI Practitioner page.
• AWS Certified Machine Learning Engineer Associate: associate, $150 for MLA-C01, 65 questions, 130 minutes, with about 1 year of experience recommended.
• AWS Certified Generative AI Developer Professional: professional, $300, 75 questions, 180 minutes, with 2 or more years of experience recommended.

The jump between each step is mostly experience, not just difficulty. The foundational exam checks that you know the terms. The associate exam checks that you’ve run machine learning workflows. The professional exam checks that you’ve built and operated generative AI applications.

How Do You Show Generative AI Experience Without an AWS Job?

Many people who want this exam don’t have an employer paying for AWS projects. You can still build the experience, and it also strengthens a résumé on its own.

Build one application you can describe. Pick a narrow problem, such as a question-answering tool over a set of documents. Run it on AWS, and write down the choices you made and why.

Track the real costs. Production work is about money as much as accuracy. Know what your application costs to run and what you’d change to lower it.

Add monitoring and a failure plan. Decide how you’d know the application is giving bad answers, and what happens when it does.

Write it up. A short README with a diagram and what you’d do differently shows more than a list of services.

A project like this helps with the exam and with job applications at once. It’s also the sort of work an online AI degree’s capstone can include, which is one reason the two fit together.

What If You Don’t Pass?

Each attempt costs the exam fee, so a failed try costs another $300 USD. Confirm AWS’s current retake policy on the certification page before you book.

If you miss, use the result to find your weakest areas, spend a few weeks on them and build another small project that touches those topics. A second attempt goes better with new hands-on work behind it than with the same notes reread.

How Does It Relate to an AI Degree?

A cert doesn’t replace a degree, and a degree doesn’t replace a cert. They measure different things.

A degree teaches machine learning theory, statistics, programming and modeling. It takes years, it gives college credit and it opens doors that require one. The AWS exam checks that you can build generative AI applications on one cloud. No employer will treat it as a degree, and AWS lists no degree requirement for taking it.

A degree also doesn’t teach you AWS by default. A graduate can finish a program without deploying anything on the platform. This exam shows you’ve done it.

Here’s how people combine them:

• Degree first, exam later. Common for early-career engineers. The degree opens the door, then two years of work and the exam show platform skill.
• Experience first, degree alongside. Working developers often take an online degree part-time while they build on AWS.
• Exam alone. It works for developers who are already employed and want a cloud-specific credential. It’s a weak path into the field from outside.

For the degree side, our ranking lists are the place to look. Online generative AI programs match this exam’s subject most closely, and online AI engineering degrees fit the build-and-deploy focus. If you’re still deciding whether a degree is worth it, read Is an AI Degree Worth It?.

What Jobs Does It Help With?

The credential points to engineering roles that build on foundation models. Our AI engineer and machine learning engineer pages cover duties, degree levels and pay for those jobs. Software engineer roles that now include generative AI features are part of the same picture.

An exam can’t get you hired. What it can do is help a hiring manager trust a line on your résumé: that you’ve worked with generative AI on AWS in production. That signal matters most at employers that run on AWS, and at consultancies that sell AWS projects.

Look at job postings in your target role. If AWS generative AI shows up in the requirements, the exam is a direct match. If it doesn’t, your projects and degree carry the weight.

Is It Worth Taking?

For the right person, yes. Consider it if:

• You build on AWS today and want a credential that matches the work.
• Your employer pays for exams or ties raises and project work to certifications.
• You’re moving into a senior generative AI role and want a clear signal on your résumé.

Skip it, or wait, if:

• You have less than a year of hands-on generative AI work. Build that first.
• You’re early in your career and need a degree. That will open more doors than the exam.
• Your target employers use a different cloud.

For people at the start, the better move is a lower rung. The AI Practitioner is $100, and the associate asks for about a year of experience. Either one will teach you what the professional exam expects.

What About the Machine Learning Specialty?

AWS retired the Machine Learning Specialty, and the last day to take it was March 31, 2026. It cost $300 and had 65 questions in 180 minutes. If you’re wondering how this exam relates to it, the two share a price and a time limit, but they cover different ground. Our page on the retired Specialty explains what to take instead.