AI-103 study plan

Week 1chapters 1–4 · about 4 hours Done

• Read these chapters:
  • 1. Microsoft Foundry: models, resources and deployments
  • 2. Securing AI solutions: identity, keys and networking
  • 3. Operating AI solutions: quotas, monitoring, evaluation and CI/CD
  • 4. Responsible AI, guardrails and Azure AI Content Safety
• After each chapter, cover its key takeaways and explain each one out loud. Re-read any you can’t.
• Free study notes for this week: Guardrails in Microsoft Foundry: intervention points, Prompt Shields and actions, Agent oversight: approvals, constraints and tool-access controls.
• Look up any unfamiliar term in the AI-103 glossary.

Exam skill areas this week: Plan and manage an Azure AI solution · 25–30%Implement generative AI and agentic solutions · 30–35%Implement text analysis solutions · 10–15%

Week 2chapters 5–9 · about 4 hours Done

• Read these chapters:
  • 5. Building generative applications: prompts, parameters and quality
  • 6. Building agents with Foundry Agent Service
  • 7. Image and video generation and multimodal understanding
  • 8. Azure Vision: image analysis and OCR
  • 9. Text analysis with Azure Language and language models
• After each chapter, cover its key takeaways and explain each one out loud. Re-read any you can’t.
• Free study notes for this week: Prompt engineering techniques for Azure AI apps, Choosing agent tools in Foundry Agent Service, Video generation with Sora 2 in Microsoft Foundry.
• Look up any unfamiliar term in the AI-103 glossary.

Exam skill areas this week: Plan and manage an Azure AI solution · 25–30%Implement generative AI and agentic solutions · 30–35%Implement computer vision solutions · 10–15%Implement text analysis solutions · 10–15%Implement information extraction solutions · 10–15%

Week 3chapters 10–13 · about 3 hours Done

• Read these chapters:
  • 10. Translation: text, documents and speech
  • 11. Speech and voice agents
  • 12. Azure AI Search and retrieval-augmented generation
  • 13. Document and content extraction
• After each chapter, cover its key takeaways and explain each one out loud. Re-read any you can’t.
• Free study notes for this week: Azure AI Search indexers, skillsets and the knowledge store, Vector, hybrid and semantic search in Azure AI Search, Azure Content Understanding: analyzers, modes and structured output.
• Look up any unfamiliar term in the AI-103 glossary.

Exam skill areas this week: Plan and manage an Azure AI solution · 25–30%Implement generative AI and agentic solutions · 30–35%Implement computer vision solutions · 10–15%Implement text analysis solutions · 10–15%Implement information extraction solutions · 10–15%

Week 4review Done

• Go back through every chapter’s key takeaways. Anything you can’t explain without looking, re-read that section.
• Re-read the common-trap callouts: they mark the answers that look right but aren’t.
• Check the exam coverage map against Microsoft’s own outline, in case it has changed since the book was checked.
• Take Microsoft’s free practice assessment, if your exam has one, from the official AI-103 study guide page. Revisit the chapters behind any weak area.
• Try the exam sandbox to get used to the exam’s screens and question formats.
• Get the free exam-day checklist and work through it a few days before your exam.

Week 1chapters 1–2 · about 2 hours Done

• Read these chapters:
  • 1. Microsoft Foundry: models, resources and deployments
  • 2. Securing AI solutions: identity, keys and networking
• After each chapter, cover its key takeaways and explain each one out loud. Re-read any you can’t.
• Look up any unfamiliar term in the AI-103 glossary.

Exam skill areas this week: Plan and manage an Azure AI solution · 25–30%Implement generative AI and agentic solutions · 30–35%

Week 2chapter 3 · about 1 hour Done

• Read chapter:
  • 3. Operating AI solutions: quotas, monitoring, evaluation and CI/CD
• After each chapter, cover its key takeaways and explain each one out loud. Re-read any you can’t.
• Look up any unfamiliar term in the AI-103 glossary.

Exam skill areas this week: Plan and manage an Azure AI solution · 25–30%Implement generative AI and agentic solutions · 30–35%

Week 3chapters 4–5 · about 2 hours Done

• Read these chapters:
  • 4. Responsible AI, guardrails and Azure AI Content Safety
  • 5. Building generative applications: prompts, parameters and quality
• After each chapter, cover its key takeaways and explain each one out loud. Re-read any you can’t.
• Free study notes for this week: Guardrails in Microsoft Foundry: intervention points, Prompt Shields and actions, Agent oversight: approvals, constraints and tool-access controls, Prompt engineering techniques for Azure AI apps.
• Look up any unfamiliar term in the AI-103 glossary.

Exam skill areas this week: Plan and manage an Azure AI solution · 25–30%Implement generative AI and agentic solutions · 30–35%Implement text analysis solutions · 10–15%Implement information extraction solutions · 10–15%

Week 4chapters 6–7 · about 2 hours Done

• Read these chapters:
  • 6. Building agents with Foundry Agent Service
  • 7. Image and video generation and multimodal understanding
• After each chapter, cover its key takeaways and explain each one out loud. Re-read any you can’t.
• Free study notes for this week: Choosing agent tools in Foundry Agent Service, Video generation with Sora 2 in Microsoft Foundry.
• Look up any unfamiliar term in the AI-103 glossary.

Exam skill areas this week: Plan and manage an Azure AI solution · 25–30%Implement generative AI and agentic solutions · 30–35%Implement computer vision solutions · 10–15%

Week 5chapters 8–9 · about 1 hour Done

• Read these chapters:
  • 8. Azure Vision: image analysis and OCR
  • 9. Text analysis with Azure Language and language models
• After each chapter, cover its key takeaways and explain each one out loud. Re-read any you can’t.
• Look up any unfamiliar term in the AI-103 glossary.

Exam skill areas this week: Implement computer vision solutions · 10–15%Implement text analysis solutions · 10–15%Implement information extraction solutions · 10–15%

Week 6chapters 10–11 · about 1 hour Done

• Read these chapters:
  • 10. Translation: text, documents and speech
  • 11. Speech and voice agents
• After each chapter, cover its key takeaways and explain each one out loud. Re-read any you can’t.
• Look up any unfamiliar term in the AI-103 glossary.

Exam skill areas this week: Implement text analysis solutions · 10–15%

Week 7chapters 12–13 · about 2 hours Done

• Read these chapters:
  • 12. Azure AI Search and retrieval-augmented generation
  • 13. Document and content extraction
• After each chapter, cover its key takeaways and explain each one out loud. Re-read any you can’t.
• Free study notes for this week: Azure AI Search indexers, skillsets and the knowledge store, Vector, hybrid and semantic search in Azure AI Search, Azure Content Understanding: analyzers, modes and structured output.
• Look up any unfamiliar term in the AI-103 glossary.

Exam skill areas this week: Plan and manage an Azure AI solution · 25–30%Implement generative AI and agentic solutions · 30–35%Implement computer vision solutions · 10–15%Implement information extraction solutions · 10–15%

Week 8review Done

• Go back through every chapter’s key takeaways. Anything you can’t explain without looking, re-read that section.
• Re-read the common-trap callouts: they mark the answers that look right but aren’t.
• Check the exam coverage map against Microsoft’s own outline, in case it has changed since the book was checked.
• Take Microsoft’s free practice assessment, if your exam has one, from the official AI-103 study guide page. Revisit the chapters behind any weak area.
• Try the exam sandbox to get used to the exam’s screens and question formats.
• Get the free exam-day checklist and work through it a few days before your exam.