AI-102 Requirements 2026: Eligibility, Prerequisites & How to Qualify

Status Change: What "AI-102 Requirements" Actually Means in 2026

If you're researching requirements for "AI-102," the first thing to understand is that the exam itself, Designing and Implementing a Microsoft Azure AI Solution, along with the Microsoft Certified: Azure AI Engineer Associate credential and its renewal assessment, all retired on June 30, 2026 at 11:59 PM Central. After that date, Microsoft stopped offering the exam entirely - it cannot be scheduled, taken, or renewed. Anyone who passed before the retirement date keeps the credential listed in the Active Certifications section of their Microsoft Learn transcript until its printed expiration, but there is no path to earn it fresh or renew it going forward.

The live successor is Exam AI-103, Developing AI Apps and Agents on Azure, which awards Microsoft Certified: Azure AI Apps and Agents Developer Associate. The associated training also moved: the four-day AI-103T00-A course replaced the older AI-102T00 course at the training level. For anyone qualifying in 2026 or later, "AI-102 requirements" functionally means AI-103 requirements - same skill area, same commercial intent, updated exam number and updated content emphasis on generative AI and agent orchestration.

Are There Official Prerequisites?

Microsoft does not enforce any formal prerequisite for this certification. There's no lower-level exam you must pass first, no mandatory course you must complete, and no application or eligibility check before you can register. Anyone who pays the exam fee and books a seat through Pearson VUE - at a physical test center or via online proctoring - can sit the exam.

That said, "no prerequisite" does not mean "no baseline expectation." Microsoft designs the item bank assuming a specific working profile, and candidates who skip that groundwork tend to struggle regardless of how much they study the exam skills outline. If you're unsure whether you're actually ready, the difficulty breakdown for the exam and the data on outcomes are worth reviewing before you commit to a registration date.

Who This Certification Assumes You Already Are

Microsoft's exam guide describes an audience profile rather than a hard requirement, but treat it as the real prerequisite. The assumed candidate has:

• Application development experience in Python, since sample scenarios and code-completion items assume comfort reading and reasoning about Python code.
• Familiarity with general AI concepts and generative AI patterns - prompting, embeddings, retrieval, and agent behavior aren't explained from scratch on the exam.
• Working exposure to Azure services generally, including how resources are provisioned, secured, and monitored inside a subscription.

If you're coming from a different stack (Node.js shops, .NET-only teams, or data science roles with limited app development background), you'll need to close that Python and Azure gap before the domain content will make sense. This is also why the certification tends to attract a specific hiring profile - teams building copilots, RAG-based search experiences, and agentic automation layers on Azure, rather than general software engineers or pure data scientists. If you're evaluating whether this fits your career direction, the roles that typically list this credential and the ROI analysis both dig into that fit in more detail.

Registration and Exam-Day Requirements

Once you've confirmed you have the background, qualifying operationally comes down to logistics rather than credentials. Here's what actually gates your attempt:

• Fee: $165 USD at standard US pricing; the amount varies by the country or region where you sit the exam.
• Delivery: Pearson VUE, either at a physical test center or through online proctoring from your own machine.
• Duration: 120 minutes, and the format may include interactive components rather than pure multiple choice.
• Passing score: 700 on a 100-1000 scale. Microsoft does not publish a fixed number of questions or a public pass rate for this exam.
• Content currency: skills measured are current as of April 16, 2026, and while most questions target generally available features, commonly used preview features can appear.

None of these are eligibility barriers in the traditional sense - you don't need to prove years of experience or submit documentation. But they do shape how you prepare: a 120-minute window with interactive items rewards people who've actually built things in Azure AI Foundry rather than memorized definitions. For a full walkthrough of exactly what "passing" requires numerically, see the passing score breakdown, and for a complete cost picture including retake fees and training, check the pricing breakdown.

The Five Domains You Have to Qualify Against

Since there's no formal prerequisite exam, the real "qualification bar" is domain coverage. The current skills outline breaks into five areas, and your readiness should be judged against each one individually rather than as a single average.

For a deeper breakdown of sub-topics and question examples inside each of these five areas, the complete domain guide is the natural next stop - it's built specifically to complement this requirements overview.

The Technology Stack You're Expected to Know

Beyond the domain outline, qualifying for this exam in practice means having built something with a specific set of Azure services. The exam expects working knowledge of:

• Microsoft Foundry and Foundry Agent Service
• Microsoft Agent Framework
• Azure OpenAI in Foundry Models
• Azure AI Search
• Azure Content Understanding in Foundry Tools
• Azure Document Intelligence in Foundry Tools
• Azure Speech
• Azure Translator

If any of these are unfamiliar names rather than tools you've actually configured, that's your real gap to close - not a certificate or course requirement. A useful self-check: try to explain, from memory, how Foundry Agent Service and Microsoft Agent Framework relate to each other, and how a tool schema gets wired into a function-calling agent. If you can't, that's exactly the territory Domain 2 tests most heavily.

Retake Rules and Annual Renewal

Two ongoing requirements matter after your first attempt:

• Retakes: a failed first attempt can be retaken after a 24-hour wait. Subsequent failed attempts carry progressively longer waiting periods, so a single miss doesn't cost you much time, but repeated misses do.
• Renewal: like other Microsoft associate certifications, this one expires annually. Renewal is free and happens by passing an unproctored online assessment on Microsoft Learn - no need to retake the full proctored exam every year.

This renewal cadence is worth planning around from day one, since it means "qualifying" for this credential is really an annual commitment rather than a one-time event. Keep an eye on your transcript date so the free renewal assessment doesn't sneak past you.

Building a Realistic Qualification Timeline

Because there's no prerequisite course to complete, most candidates default to self-study, and the order you tackle domains in matters more than the total number of hours. Given that Domain 2 carries the most weight and covers the newest, least-familiar material for most candidates, it deserves the most calendar time and should not be left for the final week.

For a more detailed week-by-week study methodology built specifically around this exam's structure, see the first-attempt study guide. And once you're in the review phase, running timed sets on the practice test platform is the fastest way to see which domains still need work before you spend $165 on the real attempt.

Frequently Asked Questions