More study resources

About this exam

AI-103 leads to the Microsoft Certified: Azure AI Apps and Agents Developer Associate credential, an intermediate-level certification for AI engineers and developers who build, manage, and deploy agents and AI solutions that take advantage of Microsoft Foundry. Candidates should already be comfortable developing apps in Python and should understand the capabilities of general AI, generative AI, and Azure services. The role spans planning and managing Azure AI solutions plus implementing generative AI and agentic, computer vision, text analysis, and information extraction workloads, and it involves collaborating with business stakeholders, solution architects, data scientists, DevOps engineers, and cloud security engineers. The assessment is proctored, lasts 120 minutes, and may include interactive components alongside other item types. It is offered in English, Chinese (Simplified), Chinese (Traditional), French, German, Japanese, Korean, Italian, Portuguese (Brazil), and Spanish, and is scheduled through Pearson VUE, with price set by the country or region where the exam is proctored. Preparation should center on the free AI-103 study guide on Microsoft Learn, reinforced with hands-on building in Microsoft Foundry, a run through the exam sandbox for interface familiarity, and the Practice Assessment available on AI Skills Navigator.

What you'll study

Common questions

About AI-103 preparation

What is the exam format?

The exam is proctored and may include interactive components that you complete as part of the assessment, alongside more conventional item types. Microsoft recommends using the exam sandbox first so you can interact with the different question types in the same user interface you will see on test day.

How long is the exam, and how many questions does it have?

You have 120 minutes to complete the assessment. The number of scored items is not published on the exam page, so plan your pacing around the interactive components, which often take longer than standard multiple-choice items.

Can I retake the exam if I fail?

Yes. You can retake it 24 hours after the first failed attempt. For subsequent retakes, the amount of time varies, so check the Microsoft exam retake policy page for the full details before you reschedule.

What experience should I have before attempting it?

The certification targets an intermediate level. You should have experience developing apps by using Python and be familiar with the capabilities of general AI, generative AI, and Azure services. Practical work in Microsoft Foundry, rather than reading alone, is the best way to build that familiarity.

In which languages is the exam offered?

This exam is offered in English, Chinese (Simplified), Chinese (Traditional), French, German, Japanese, Korean, Italian, Portuguese (Brazil), and Spanish. Remember that the language you select affects the wording of scenario questions, so practice with technical vocabulary in your chosen language.

How should I order my study time?

Start with the official AI-103 study guide to see the measured skill areas and any listed updates, then work through hands-on exercises in Microsoft Foundry for each area, prioritizing the ones you use least at work. Finish with the Practice Assessment on AI Skills Navigator and the exam sandbox.

What is the passing score?

The exam page does not publish a passing score, so this guide cannot state one. Treat any figure you see elsewhere as unverified and focus instead on the study guide's skill outline, which is the documented basis for what the exam measures.

How much does the exam cost?

Pricing is based on the country or region in which the exam is proctored, and the exam page gives no single figure. Check the current price for your location when you schedule through Pearson VUE, since it can differ from another candidate's in a different market.

How should I register for the exam?

Microsoft strongly recommends registering with a personal MSA account. If you register with an organizational work or school account, your exam records will be lost if you leave that organization, and they will be unrecoverable.

Why do capable engineers still fail?

The usual reason is preparation limited to reading documentation. Candidates who never build agents, wire up AI services in Python, or work through Foundry configuration struggle with scenario items. Practice the full build-and-deploy cycle, and rehearse the exam sandbox so unfamiliar item types do not cost you time.