examOS.
Exam CatalogueStudy PlansRoadmapsBlogs
Login

ExamOS

Credits PolicyReferral PolicyQuality StandardsPricingPrivacy PolicyTerms of UseContact UsReport a Bug

Follow us

Disclaimer: ExamOS is an independent platform, not affiliated with any certification provider, and does not use or distribute exam dumps.

Share your feedback

Checking sign-in status...

examOS.Study Plan
Disclaimer: ExamOS is an independent platform, not affiliated with any certification provider, and does not use or distribute exam dumps.
← Back to Exam Details
examOS.Study Plan · examos.io

Study Plan

Azure AI Fundamentals (AI-901) Study Plan for 2026

A 4-week plan for AI-901 covering AI concepts, responsible AI, Microsoft Foundry, generative AI apps, agents, text, speech, vision, and information extraction.

MicrosoftAI-901Passing score: 700 / 1000Beginners with basic Python familiarity and no prior AI certification, ready to build foundational AI skills in Microsoft Foundry14-Sep-202624 views
Start date: _______________Target exam date: _______________

Your personalized plan

A focused route built around your experience and daily study commitment.

Timeline

4 weeks

Personalized calendar

Daily commitment

1h

5 study & practice hrs/week

Progress

0/4

milestones completed

Saved on this device

Your week-by-week breakdown


Your week 1

Milestone 1: AI Concepts and Responsible AI (Domain 1)

This week covers the conceptual foundation: what AI workloads exist, how generative models work, and the six responsible AI principles. Everything in Domain 2 assumes you understand this vocabulary.

Topics

  • • Responsible AI: fairness, reliability and safety, privacy and security, inclusiveness, transparency, accountability
  • • Generative AI models: how they work, model selection, deployment options
  • • AI workload families: generative AI, agentic AI, text analysis, speech, computer vision, information extraction
  • • Common text analysis techniques: keyword extraction, entity detection, sentiment analysis, summarization
  • • Speech recognition and synthesis
  • • Computer vision and image generation capabilities

Activities

  • • Complete all six modules of the Microsoft Learn path AI concepts for developers and technology professionals.
  • • Write the six responsible AI principles from memory. For each one, write one sentence explaining what it means in practice.
  • • Match five sample scenarios to the correct AI workload family (text, speech, vision, extraction, generative, agentic).
  • • Visit the exam sandbox at `aka.ms/examdemo` and try at least two sample questions.
  • • Complete a Daily Sprint in Rookie mode. Flag any topic below 60%.

Your week 2

Milestone 2: Foundry Foundations, Generative AI, and Agents (Domain 2)

This is where the exam stops being theoretical. You deploy your first model, interact with it in the Microsoft Foundry portal, build a lightweight chat client with the Foundry SDK, and create a single-agent solution. This week carries the most weight in Domain 2.

Topics

  • • Microsoft Foundry portal: projects, model deployment, playground
  • • Foundry Tools overview: Speech, Vision, Language, Content Safety, Content Understanding
  • • System prompts and user prompts
  • • Foundry SDK: creating a lightweight chat client
  • • Agent fundamentals: creating and testing a single-agent solution
  • • Calling an agent from application code

Activities

  • • Complete the Microsoft Learn modules Get started with AI in Azure and Get started with generative AI and agents in Azure.
  • • Deploy a model in the Foundry portal. Interact with it in the playground. Write two system prompts and two user prompts. Note how the output changes.
  • • Build a lightweight chat client using the Foundry SDK. The pattern is: create an `AIProjectClient` with your project endpoint and credential, request an OpenAI-compatible client, send input, print the response.
  • • Create a single-agent solution in the Foundry portal. Test it with at least three different prompts.
  • • Build a lightweight client application that calls your agent.
  • • Complete a Daily Sprint in Challenger mode. Repeat until you pass two consecutive sessions.

Your week 3

Milestone 3: Text, Speech, and Computer Vision (Domain 2)

This week covers the middle two hands-on blocks of Domain 2. You build applications that analyze text, respond to spoken prompts, and interpret visual input. The exam tests whether you understand what each Foundry Tool does and when to use it.

Topics

  • • Text analysis in Foundry Tools: keyword extraction, entity detection, sentiment, summarization
  • • Azure Speech in Foundry Tools: speech recognition, speech synthesis
  • • Computer vision: interpreting visual input with a multimodal model
  • • Image generation: creating new visual outputs with generative models
  • • Building lightweight applications for text, speech, and vision

Activities

  • • Complete the Microsoft Learn modules Get started with text analysis in Azure and Get started with speech in Azure.
  • • Build a small Python client that analyzes text for sentiment and extracts entities.
  • • Build a lightweight application that uses Azure Speech in Foundry Tools to recognize and synthesize speech.
  • • Use a deployed multimodal model to interpret an image and answer a question about it.
  • • Generate an image from a text prompt using a generative model.
  • • Complete a Daily Sprint in Challenger mode. Repeat until you pass two consecutive sessions.

Your week 4

Milestone 4: Information Extraction and Exam Simulation (Domain 2 + Review)

The final week covers information extraction using Azure Content Understanding, then shifts to full exam simulation. You have built the pieces. Now consolidate.

Topics

  • • Azure Content Understanding in Foundry Tools
  • • Extracting information from documents, forms, images, audio, and video
  • • Building a lightweight application with information extraction
  • • Full syllabus review across both domains
  • • Time management (45-65 minutes for 40-60 questions)
  • • Scenario reasoning and distractor elimination

Activities

  • • Complete the Microsoft Learn module Get started with AI-powered information extraction in Azure.
  • • Use Azure Content Understanding to extract structured data from a document and from an image.
  • • Build a lightweight application that extracts information from a sample document.
  • • Take the free official practice assessment on AI Skills Navigator. Compare your score to Week 1.
  • • Take at least 3 Mock Exams in Legend mode. For every missed question, write one sentence explaining the principle behind the correct answer.
  • • Simulate real exam conditions once: no phone, no breaks, timed, same time of day as your scheduled exam.
  • • Book your exam when your readiness score is stable and above 80, with every domain covered.

Stay consistent by setting a target date for this certification.

Set target

Week-by-Week Breakdown


W1

Week 1

AI Concepts and Responsible AI (Domain 1)

This week covers the conceptual foundation: what AI workloads exist, how generative models work, and the six responsible AI principles. Everything in Domain 2 assumes you understand this vocabulary.

Topics

  • Responsible AI: fairness, reliability and safety, privacy and security, inclusiveness, transparency, accountability
  • Generative AI models: how they work, model selection, deployment options
  • AI workload families: generative AI, agentic AI, text analysis, speech, computer vision, information extraction
  • Common text analysis techniques: keyword extraction, entity detection, sentiment analysis, summarization
  • Speech recognition and synthesis

Daily Study Routine

Suggested 2–3 Hour Day

TimeActivity
10 minReview yesterday's missed questions. Identify the principle you missed.
30 minWork through the Microsoft Learn module for this week's topic.
30 minHands-on lab in Microsoft Foundry portal or with the Foundry SDK.
20 minComplete a Daily Sprint or Mock Exam.
  • Computer vision and image generation capabilities
  • Activities

    • Complete all six modules of the Microsoft Learn path AI concepts for developers and technology professionals.
    • Write the six responsible AI principles from memory. For each one, write one sentence explaining what it means in practice.
    • Match five sample scenarios to the correct AI workload family (text, speech, vision, extraction, generative, agentic).
    • Visit the exam sandbox at `aka.ms/examdemo` and try at least two sample questions.
    • Complete a Daily Sprint in Rookie mode. Flag any topic below 60%.
    W2

    Week 2

    Foundry Foundations, Generative AI, and Agents (Domain 2)

    This is where the exam stops being theoretical. You deploy your first model, interact with it in the Microsoft Foundry portal, build a lightweight chat client with the Foundry SDK, and create a single-agent solution. This week carries the most weight in Domain 2.

    Topics

    • Microsoft Foundry portal: projects, model deployment, playground
    • Foundry Tools overview: Speech, Vision, Language, Content Safety, Content Understanding
    • System prompts and user prompts
    • Foundry SDK: creating a lightweight chat client
    • Agent fundamentals: creating and testing a single-agent solution
    • Calling an agent from application code

    Activities

    • Complete the Microsoft Learn modules Get started with AI in Azure and Get started with generative AI and agents in Azure.
    • Deploy a model in the Foundry portal. Interact with it in the playground. Write two system prompts and two user prompts. Note how the output changes.
    • Build a lightweight chat client using the Foundry SDK. The pattern is: create an `AIProjectClient` with your project endpoint and credential, request an OpenAI-compatible client, send input, print the response.
    • Create a single-agent solution in the Foundry portal. Test it with at least three different prompts.
    • Build a lightweight client application that calls your agent.
    • Complete a Daily Sprint in Challenger mode. Repeat until you pass two consecutive sessions.
    W3

    Week 3

    Text, Speech, and Computer Vision (Domain 2)

    This week covers the middle two hands-on blocks of Domain 2. You build applications that analyze text, respond to spoken prompts, and interpret visual input. The exam tests whether you understand what each Foundry Tool does and when to use it.

    Topics

    • Text analysis in Foundry Tools: keyword extraction, entity detection, sentiment, summarization
    • Azure Speech in Foundry Tools: speech recognition, speech synthesis
    • Computer vision: interpreting visual input with a multimodal model
    • Image generation: creating new visual outputs with generative models
    • Building lightweight applications for text, speech, and vision

    Activities

    • Complete the Microsoft Learn modules Get started with text analysis in Azure and Get started with speech in Azure.
    • Build a small Python client that analyzes text for sentiment and extracts entities.
    • Build a lightweight application that uses Azure Speech in Foundry Tools to recognize and synthesize speech.
    • Use a deployed multimodal model to interpret an image and answer a question about it.
    • Generate an image from a text prompt using a generative model.
    • Complete a Daily Sprint in Challenger mode. Repeat until you pass two consecutive sessions.
    W4

    Week 4

    Information Extraction and Exam Simulation (Domain 2 + Review)

    The final week covers information extraction using Azure Content Understanding, then shifts to full exam simulation. You have built the pieces. Now consolidate.

    Topics

    • Azure Content Understanding in Foundry Tools
    • Extracting information from documents, forms, images, audio, and video
    • Building a lightweight application with information extraction
    • Full syllabus review across both domains
    • Time management (45-65 minutes for 40-60 questions)
    • Scenario reasoning and distractor elimination

    Activities

    • Complete the Microsoft Learn module Get started with AI-powered information extraction in Azure.
    • Use Azure Content Understanding to extract structured data from a document and from an image.
    • Build a lightweight application that extracts information from a sample document.
    • Take the free official practice assessment on AI Skills Navigator. Compare your score to Week 1.
    • Take at least 3 Mock Exams in Legend mode. For every missed question, write one sentence explaining the principle behind the correct answer.
    • Simulate real exam conditions once: no phone, no breaks, timed, same time of day as your scheduled exam.
    • Book your exam when your readiness score is stable and above 80, with every domain covered.

    Overview

    AI-901 replaced AI-900 on June 30, 2026. The credential name stays the same, but the exam is meaningfully different. AI-900 tested recognition. AI-901 tests whether you can build something in Microsoft Foundry, not just name the services.

    The exam has two domains, and the larger one is entirely hands-on. More than half your score depends on whether you have actually deployed a model, created an agent, and built a lightweight application in Foundry. Reading alone will not get you there.

    Domain Weight
    Identify AI concepts and capabilities 40-45%
    Implement AI solutions by using Microsoft Foundry 55-60%

    Domain 1 covers responsible AI, model components, and AI workload families. Domain 2 is four hands-on blocks: generative AI apps and agents, text and speech, computer vision and image generation, and information extraction.

    Recommended experience: Basic Python familiarity. You do not need prior AI experience. You need to be comfortable reading a short Python script, creating a resource in the Azure portal, and understanding what an API call does. If any of those are unfamiliar, spend a few hours on Python basics before Week 1.

    A note on prerequisites: There are no formal prerequisites. If you are new to Azure, create a free account before starting. The exam expects familiarity with Azure resources, REST APIs, SDKs, and CLIs at a conceptual level.

    FAQ

    Do I need AI-900 before AI-901?

    No. AI-900 retired on June 30, 2026. AI-901 is the only path to the Azure AI Fundamentals credential. If you already passed AI-900, your certification remains valid for life and you do not need to retake anything.

    How long does it take to prepare for AI-901?

    Two to four weeks for complete beginners who can study one to one and a half hours per day. One to two weeks if you already work in IT or cloud and are comfortable with Python and Azure resources.

    Do I need to know Python?

    Yes, at a foundational level. The exam expects knowledge of Python coding syntax and programming techniques. You need to read and modify short scripts that use the Foundry SDK. You do not need to write production applications from scratch.

    What is Microsoft Foundry?

    Microsoft Foundry is the unified platform for building AI applications and agents on Azure. It consolidated what used to be scattered across Azure OpenAI Service and Azure AI Studio. The current terminology uses Microsoft Foundry for the platform and Foundry Tools for the AI services portfolio (Speech, Vision, Language, Content Safety, Content Understanding).

    What are Foundry Tools?

    Foundry Tools is the current name for the AI services portfolio that was previously called Azure AI Services (and before that, Azure Cognitive Services). It includes Speech, Vision, Language, Content Safety, and Content Understanding.

    How many practice questions should I aim for?

    At least 400 unique questions across both domains. Focus on scenario-based questions that test application, not recall. ExamOS provides practice questions designed to explain the rationale behind correct and incorrect answers.

    Can I use Microsoft Learn during the exam?

    No. Fundamentals exams do not provide access to Microsoft Learn during the test. You must answer from your knowledge.

    What is the most important domain to study?

    Implement AI solutions by using Microsoft Foundry (55-60%) is the larger domain and the one that requires hands-on practice. Domain 1 (40-45%) provides the vocabulary and responsible AI foundation. Do not skip Domain 1, but budget more time for Domain 2.

    What are the most common ways people fail?

    • Reading without building. AI-901 tests whether you can deploy a model, create an agent, and build a lightweight application. If you have not done these things, you will struggle with Domain 2 questions.
    • Skipping responsible AI. The six principles appear across both domains and are reliable marks if you learn them properly.
    • Confusing Foundry terminology. The platform went through name changes. Current terms are Microsoft Foundry and Foundry Tools. Older materials may say Azure AI Foundry, Azure AI Studio, or Azure Cognitive Services.
    • Not understanding the four hands-on blocks. Domain 2 splits into generative AI and agents, text and speech, computer vision and image generation, and information extraction. Each is roughly equal in weight.

    What are the three biggest preparation mistakes?

    1. Studying without touching Microsoft Foundry. The exam is implementation-heavy. Deploy a model. Build a chat client. Create an agent. These are not optional.
    2. Ignoring information extraction. Azure Content Understanding is a distinct block within Domain 2. Do not leave it to the last day.
    3. Not practicing under time pressure. 40-60 questions in 45-65 minutes means roughly one minute per question. Scenario questions may take longer. Practice timed Mock Exams.

    Is there a renewal option?

    No. Microsoft Fundamentals certifications do not expire and do not require renewal. Once you pass AI-901, the credential is yours for life.