AI-103 Study HubDeveloping AI Apps and Agents on Azure

Notes, interactive quizzes, flashcards, and a full timed mock exam — covering all 5 exam domains.

Domain 1 · Plan & manage AI

Choose Foundry models & services, set up deployments & CI/CD, manage quotas/monitoring/security, and apply responsible AI. (25–30%)

🤖

Domain 2 · Generative AI & agents

Build generative apps and RAG, design agents with tools/memory, orchestrate multi-agent flows, and operationalize with tracing. (30–35%)

👁️

Domain 3 · Computer vision

Image & video generation/editing, multimodal understanding and captioning, and responsible AI for visual content. (10–15%)

📝

Domain 4 · Text analysis

Entity/sentiment/summary extraction, translation, and speech-to-text / text-to-speech for agentic interactions. (10–15%)

📄

Domain 5 · Information extraction

Retrieval & grounding pipelines, semantic/hybrid/vector search, OCR + layout, and Content Understanding document extraction. (10–15%)

🧠

Quizzes & flashcards

Per-domain single-answer quizzes with instant explanations, plus rapid-recall flashcards.

☑️

Multiple-response

Exam-style "select all that apply" (pick-2/pick-3) questions with all-or-nothing scoring.

🧩

Exam formats

Interactive build-list, hotspot-dropdown, and matching items — the question types beyond plain multiple choice.

📋

Case studies

Exam-style multi-question scenarios with a persistent scenario panel and linked questions.

🔁

Problem-solution sets

One goal, several proposed solutions, each judged Yes/No — and no going back once answered.

⏱️

Timed mock exam

Sampled, 120-minute exam with a case-study section, multiple-response items, palette, and per-domain breakdown.

📄

Cram sheet

Every flashcard fact condensed onto one printable page, grouped by domain — for the final hour.

🎯

Review weak spots

Every question you miss anywhere is tracked, so you can drill exactly what you got wrong.

📈

My progress

A dashboard of your saved quiz, flashcard, case-study, and mock-exam scores — all on-device.

How to use this site

AI-103 blueprint — five domains

#	Domain	Weight
1	Plan and manage an Azure AI solution	25–30%
2	Implement generative AI and agentic solutions	30–35%
3	Implement computer vision solutions	10–15%
4	Implement text analysis solutions	10–15%
5	Implement information extraction solutions	10–15%

A pass on the real exam requires a scaled score of 700/1000. Work through the domain notes, test yourself with the per-domain quizzes, drill the flashcards until recall is automatic, then sit the timed mock exam to simulate the real thing.

Note: This is an unofficial study aid. Content is based on the published Microsoft Learn AI-103 study guide. Always confirm details against current Microsoft documentation.