AI-102 Cheat Sheet 2026: One-Page Review of Must-Know Facts

Status Check: What "AI-102" Means in 2026

Before you memorize a single acronym, get the timeline straight. Exam AI-102, Designing and Implementing a Microsoft Azure AI Solution, along with the credential Microsoft Certified: Azure AI Engineer Associate, retired on June 30, 2026 at 11:59 PM Central. The renewal assessment tied to that credential retired with it. That means AI-102 can no longer be scheduled, taken, or renewed by anyone starting from scratch after that date.

If you already held the credential before retirement, it stays listed in the Active Certifications section of your Microsoft Learn transcript until its printed expiration date - but there is no renewal path once that date passes. The commercially and technically correct move for anyone searching "AI-102" today is to prepare for its live successor, Exam AI-103, Developing AI Apps and Agents on Azure, which awards Microsoft Certified: Azure AI Apps and Agents Developer Associate. The training line moved too: AI-103T00-A, a four-day course, replaced AI-102T00 at the training level.

Exam Mechanics at a Glance

This is the section to screenshot. It compresses the logistics you need before you book a seat.

• Delivery: Pearson VUE, at a test center or via online proctoring.
• Duration: 120 minutes, proctored, and the exam may include interactive components rather than pure multiple choice.
• Passing score: 700 out of a 100-1000 scale. Microsoft does not publish a fixed item count or a pass rate for this exam.
• Price: $165 USD at standard US pricing; the amount varies depending on the country or region where you sit the exam.
• Prerequisites: None formally required, though the audience profile assumes app development experience in Python plus familiarity with general AI, generative AI, and Azure services.
• Currency of content: Skills measured are current as of April 16, 2026. Most questions target generally available features, though commonly used preview features can appear.

For a granular breakdown of exactly what 700 represents and how scaled scoring works, read the AI-102 Passing Score 2026 guide. If budgeting matters to your plan, the AI-102 Certification Cost breakdown covers regional pricing nuances in more depth than this cheat sheet has room for.

The Five Domains, Fast

AI-102 content is organized into five measured domains. Weightings tell you where to spend hours, not just what to read.

Want the long-form version of each bullet above, with subtopic lists and study sequencing? That lives in the AI-102 Exam Domains 2026 guide, which pairs well with this cheat sheet as a working reference.

Domain 2 Deep Dive: Generative AI and Agentic Solutions

Because this domain carries 30-35% of the exam - more than any other single band - it deserves its own section even on a cheat sheet. Candidates consistently underestimate how much of the exam has shifted toward agent design rather than static prompt engineering.

• RAG implementation: chunking strategy, grounding data selection, and retrieval quality tradeoffs.
• Multi-agent orchestration: when to split a workflow across agents versus keep it single-agent, and how handoffs are structured.
• Function calling and tool schemas: defining parameters and return contracts an agent can call deterministically.
• Conversation memory: short-term versus persisted memory and how it affects agent behavior over long sessions.
• Agent evaluation and error analysis: reading evaluation output to identify failure modes before shipping.
• Tracing and token analytics: using trace data to debug agent decisions and monitor token consumption.
• Autonomous workflows with approval controls: designing human-in-the-loop checkpoints for actions an agent shouldn't take unsupervised.

If you're still deciding how much time this domain deserves relative to the other four, the How Hard Is the AI-102 Exam? guide breaks down where most candidates report friction.

Tools and Services You Must Recognize

The exam expects working knowledge of a specific technology stack, not vague AI familiarity. Know what each one does and where it fits in an architecture diagram:

• Microsoft Foundry and Foundry Agent Service - the platform and runtime for building and hosting agents.
• Microsoft Agent Framework - the development framework for orchestrating agent logic.
• Azure OpenAI in Foundry Models - model access and deployment inside Foundry.
• Azure AI Search - the retrieval layer behind most RAG implementations.
• Azure Content Understanding in Foundry Tools and Azure Document Intelligence in Foundry Tools - extraction and multimodal content processing, tied closely to Domain 5.
• Azure Speech and Azure Translator - voice and language services that surface in text and vision scenario questions alike.

Registration, Fees, and Retake Rules

Registration runs through Pearson VUE. Standard US pricing is $165 USD, though the figure shifts based on the country or region where you sit the exam - always confirm local pricing at booking rather than assuming the US rate applies.

If your first attempt doesn't clear 700, you can retake after a 24-hour wait. Subsequent failed attempts carry progressively longer waiting periods, so treat the first sitting seriously rather than as a "free look." There's no cap mentioned on total attempts, but each retake still costs full price.

Renewal Without Retesting

Once earned, Microsoft associate certifications - including this one - expire annually. Renewal doesn't require walking back into a Pearson VUE center. Instead, you pass a free, unproctored online assessment through Microsoft Learn before the expiration date. Set a calendar reminder roughly a month ahead so a lapsed certification doesn't quietly fall off your active transcript.

One Week Per Domain: A Fast-Reference Schedule

If you're building a study calendar around this cheat sheet, weight your weeks the same way Microsoft weights the exam. Generic study techniques - spaced repetition, timed practice blocks - only matter if you're pointing them at the right domain at the right time.

For a more detailed week-by-week narrative with daily task breakdowns, see the full AI-102 Study Guide 2026. When you're ready to test retention under real exam conditions, run timed sets on the AI-102 practice test platform rather than passively re-reading notes.

Quick Comparison Table

Curious how this maps to actual hiring demand? The AI-102 Jobs overview and AI-102 Salary Guide look at who's hiring for these skills and how the credential factors into compensation conversations. If you're still weighing whether to invest the study hours at all, Is the AI-102 Certification Worth It? walks through the ROI case directly.

FAQ