AI-102: Azure AI Engineer
Exam version: Skills measured as of May 2025 | Passing score: 700/1000 | Duration: ~100-120 minutes | Cost: ~$20-50 for labs
This exam retires on June 30, 2026. Plan your preparation timeline accordingly. Microsoft may replace it with an updated certification aligned to Azure AI Foundry.
Who is this for?
The AI-102 validates the ability to design, build, and deploy AI solutions using Azure AI services. This is a hands-on, code-first certification designed for:
- AI developers building intelligent applications
- Data scientists transitioning to production AI engineering
- Software engineers integrating AI/ML into applications
- Cloud architects designing AI-powered solutions
- DevOps engineers deploying and managing AI workloads
Skills at a glance
| Domain | Weight | Challenges |
|---|---|---|
| Plan and manage an Azure AI solution | 20–25% | 01–10 |
| Implement generative AI solutions | 15–20% | 11–20 |
| Implement AI agent solutions | 5–10% | 21–23 |
| Implement computer vision solutions | 10–15% | 24–30 |
| Implement natural language processing solutions | 15–20% | 31–39 |
| Implement knowledge mining and document intelligence solutions | 15–20% | 40–49 |
All 49 challenges are code-first with Python / C# / REST tabs. Each challenge includes deploying real Azure AI resources, writing integration code, and validating results programmatically. Expect to work with SDKs, REST APIs, and Azure AI Studio.
How to use this prep
Each challenge follows a developer-focused format:
| Section | Purpose |
|---|---|
| Exam Skills Covered | Official skills this challenge addresses |
| Scenario | Real-world problem that frames the challenge |
| Architecture | Diagram of the solution components |
| Tasks | Step-by-step with Python / C# / REST tabs |
| Validation | Programmatic checks to verify your solution |
| Break & Fix | Deliberate misconfigurations to troubleshoot |
| Knowledge Check | Exam-style questions |
| Cleanup | Scripts to delete resources and avoid costs |
Prerequisites
- Programming experience — Python or C# (most challenges provide both)
- Azure subscription — Azure Free Account with $200 credit recommended
- Development environment — VS Code with Azure extensions, or GitHub Codespaces
- Azure CLI — Authenticated and configured
- AI-900 knowledge — Understanding of fundamental AI concepts (recommended, not required)
Some challenges create billable resources (Azure OpenAI, Azure AI Search, Azure AI Document Intelligence). Always run cleanup scripts after completing a challenge. Estimated total cost for all labs: $20-50 with careful cleanup.
Study resources
| Resource | Link |
|---|---|
| Official Study Guide | AI-102 Study Guide |
| Free Practice Assessment | Practice Questions |
| Exam Sandbox | Try the exam interface |
| Schedule the Exam | Pearson VUE |
Microsoft Learn paths
Ready? Start with Challenge 01 to plan and provision your first Azure AI solution.