Microsoft Certified: Azure AI Engineer Associate (AI-102) Exam Blueprint

Quick Facts

Cost

Varies by country/region

Duration

100 minutes (exam time)

Questions

Not disclosed by Microsoft

Passing score

700 / 1000 (scaled)

Languages

English, Japanese, Simplified/Traditional Chinese, Korean, German, French, Spanish, Portuguese (BR), Italian

Delivery

Pearson VUE - test center OR OnVUE online proctored

Validity

1 year (free renewal via Microsoft Learn until retirement)

Retake wait

24 hr first retake, 14 days subsequent, max 5/12 months

Skills You'll Learn

Plan, create, and deploy Microsoft Foundry Services for production AI workloads. Build generative AI solutions with Microsoft Foundry, Azure OpenAI, and the Foundry SDK. Implement RAG by grounding models in your data using Foundry hubs, projects, and prompt flows. Build agentic solutions with Microsoft Foundry Agent Service and Microsoft Agent Framework. Develop computer vision solutions for image analysis, custom vision, and video insights. Implement NLP solutions for sentiment, entities, translation, speech, custom language models. Build knowledge mining with Azure AI Search, Document Intelligence, Content Understanding. Apply Responsible AI - content moderation, prompt shields, harm detection, content filters. Secure, monitor, and manage AI resources with cost controls, auth, and CI/CD pipelines. Earn the Azure AI Engineer Associate credential (retiring June 30, 2026).

Exam Logistics

Question types not enumerated on the exam page; exam may have interactive components. Use exam sandbox to preview.
Last skills-measured update was December 23, 2025 ('Microsoft Foundry' rebrand).
Exam page may show 'No training available' but study guide still references AI-102T00.
If preferred language unavailable, request an additional 30 minutes.

Prerequisites & Recommended Experience

AI-900 is not a formal prerequisite (covers foundational concepts).
Experience developing solutions in Python and C#.
Use REST APIs and SDKs to build secure image, video, NLP, knowledge mining, and GenAI solutions on Azure.
Understanding of Azure AI portfolio and storage options.
Apply Responsible AI principles.

Skills Measured with Weights (as of December 23, 2025)

Plan and manage an Azure AI solution 20-25%

Select appropriate Foundry Services for GenAI, computer vision, NLP, speech, info extraction, knowledge mining.
Plan, create, deploy a Foundry Service (Responsible AI, model selection, SDKs/APIs, CI/CD, containers).
Manage, monitor, secure (monitoring, cost, key protection, auth).
Implement AI responsibly (content moderation, content safety, prompt shields, harm detection).

Implement generative AI solutions 15-20%

Build with Microsoft Foundry (hub/project, prompt flow, RAG, SDK integration).
Azure OpenAI in Foundry Models (provision, deploy, submit prompts, DALL-E, multimodal).
Optimize and operationalize (parameters, monitoring, scalability, tracing, prompt engineering, fine-tuning).

Implement an agentic solution 5-10%

Configure resources for an agent.
Create agents with Foundry Agent Service.
Implement complex agents with Microsoft Agent Framework.
Multi-agent workflows; test, optimize, deploy.

Implement computer vision solutions 10-15%

Analyze images (visual features, object detection, OCR via Azure Vision).
Custom vision models (classification, detection, labeling, training, evaluation).
Analyze videos (Video Indexer; Spatial Analysis).

Implement natural language processing solutions 15-20%

Analyze and translate text (key phrases, entities, sentiment, PII, Translator).
Process and translate speech (TTS/STT, SSML, custom speech).
Custom language models (intents, entities, training, knowledge bases, multi-language).

Implement knowledge mining and information extraction solutions 15-20%

Azure AI Search (index, skillset, custom skills, semantic and vector).
Azure Document Intelligence (prebuilt and custom models).
Azure Content Understanding (OCR pipelines, summarize, classify, extract entities).

Official Prep Resources

Official AI-102 exam page Official Study Guide Free Practice Assessment Exam sandbox AI-102T00 instructor-led course Azure AI services docs

Test what you've learned

Take a free GoLearnQuiz practice test. Sign in to save your score.

Beginner Warm-up and concepts Start Intermediate Real exam style Start Advanced Tough scenarios Start

Additional Helpful Details

Plan retirement carefully - cert and exam end June 30, 2026.
Skills-measured rebranded recently - Microsoft Foundry naming throughout.

Microsoft Certified: Azure AI Engineer Associate (AI-102)