Microsoft Learn

Hands-on labs for AI-901

Official Microsoft Learn exercises mapped to the exam. Easy labs run in the browser. Foundry labs are the ones Microsoft hosts on a virtual machine in class — for self-study, open the module and launch the exercise from there. A free Azure account is enough.

Labs listed

16

All from Microsoft Learn

Finished

—

Tracked on this device

Foundry share of the exam

55–60%

9 portal labs · 7 easy

Official learning path

AI concepts for developers and technology professionals

The official beginner path for domain 1. Seven modules, including the easy in-browser exercises below.

Open on Microsoft Learn

3 hr 51 min

Official learning path

Get started with AI applications and agents on Azure

The official Foundry path for domain 2. Same labs Microsoft hosts on a virtual machine in the AI-901 classroom.

Open on Microsoft Learn

Beginner · Microsoft Foundry

  1. Explore AI workloads

    15 min

    Open a simple in-browser AI app and sort generative, agentic, vision, speech, and extraction scenarios. No Azure account.

    Introduction to AI concepts·Runs in the browser·Identify AI concepts (40–45%)·Episode 04

    • AI workloads
    • Generative and agentic AI
  2. Explore generative AI

    15 min

    Use a chat playground to send prompts, watch a model reply, and feel the difference a system instruction makes.

    Introduction to generative AI and agents·Runs in the browser·Identify AI concepts (40–45%)·Episode 03

    • Prompts
    • LLMs
    • Agents
  3. Explore text analytics

    15 min

    Run key-phrase, entity, sentiment, and summarization tasks on sample text so the NLP vocabulary sticks.

    Introduction to natural language processing concepts·Runs in the browser·Identify AI concepts (40–45%)·Episode 05

    • Keyword extraction
    • Entities
    • Sentiment
    • Summarization
  4. Explore AI speech

    15 min

    Try speech-to-text and text-to-speech with a generative model. Recognition versus synthesis, in fifteen minutes.

    Introduction to AI speech concepts·Runs in the browser·Identify AI concepts (40–45%)·Episode 05

    • Speech recognition
    • Speech synthesis
  5. Explore computer vision

    15 min

    Analyze an image with a multimodal model, then generate a new one. Covers the vision side of the exam map.

    Introduction to computer vision concepts·Runs in the browser·Identify AI concepts (40–45%)·Episode 06

    • Image analysis
    • Image generation
  6. Explore information extraction

    15 min

    Pull fields from a document with OCR plus generative AI. The conceptual twin of Content Understanding.

    Introduction to AI-powered information extraction concepts·Runs in the browser·Identify AI concepts (40–45%)·Episode 06

    • OCR
    • Forms
    • Field extraction
  7. Explore retrieval-augmented generation

    15 min

    Ground a chat prompt in a knowledge source and see why RAG answers stay on the documents you give it.

    Introduction to retrieval-augmented generation concepts·Runs in the browser·Identify AI concepts (40–45%)·Episode 04

    • RAG
    • Grounding
    • Knowledge
  8. Explore AI in Microsoft Foundry

    20 min

    Create a Foundry project, open the portal, and take a first walk through models, playgrounds, and agents.

    Get started with AI in Microsoft Foundry·Hosted VM in class · Azure for self-paced·Implement with Foundry (55–60%)·Episode 07

    • Foundry project
    • Portal
    • Model catalog
  9. Get started with AI in Azure

    30 min

    The official AI-901 module for spinning up Foundry, deploying a first model, and connecting it to an app.

    Get started with AI in Azure·Hosted VM in class · Azure for self-paced·Implement with Foundry (55–60%)·Episode 07

    • Azure resources
    • Foundry project
    • Deploy a model
  10. Get started with generative AI and agents

    35 min

    Deploy a model, test prompts in the Playground, then create and call a single-agent solution — the heart of the exam.

    Get started with generative AI and agents in Azure·Hosted VM in class · Azure for self-paced·Implement with Foundry (55–60%)·Episode 08

    • System and user prompts
    • Model playground
    • Single-agent solution
  11. Explore, deploy, and chat with language models

    25 min

    Pick a model from the catalog, deploy it, and chat with it in the Foundry portal before you write any code.

    Explore and deploy models from the model catalog·Hosted VM in class · Azure for self-paced·Implement with Foundry (55–60%)·Episode 07

    • Model catalog
    • Deploy a model
    • Playground chat
  12. Build a lightweight chat client with the Foundry SDK

    25 min

    The exam wants Python: AIProjectClient, an endpoint, and a short chat loop. This is that module.

    Develop an AI app with the Microsoft Foundry SDK·Hosted VM in class · Azure for self-paced·Implement with Foundry (55–60%)·Episode 07

    • Foundry SDK
    • Chat client
    • Python
  13. Get started with text analysis in Azure

    25 min

    Use general-purpose models and Azure Language in Foundry Tools, then build a lightweight Python text-analysis client.

    Get started with text analysis in Azure·Hosted VM in class · Azure for self-paced·Implement with Foundry (55–60%)·Episode 09

    • Text analysis
    • Azure Language
    • Lightweight app
  14. Get started with speech in Azure

    25 min

    Speech-to-text, text-to-speech, and Voice Live inside Foundry Tools. Matches the spoken-prompt exam skill.

    Get started with speech in Azure·Hosted VM in class · Azure for self-paced·Implement with Foundry (55–60%)·Episode 09

    • Speech-to-text
    • Text-to-speech
    • Voice Live
  15. Get started with computer vision in Azure

    30 min

    Interpret visual input with a multimodal model and generate new images — both listed on the skills measured.

    Get started with computer vision in Azure·Hosted VM in class · Azure for self-paced·Implement with Foundry (55–60%)·Episode 09

    • Multimodal vision
    • Image generation
    • Lightweight app
  16. Get started with Content Understanding

    25 min

    Extract structured fields from documents, images, audio, and video with Azure Content Understanding in Foundry Tools.

    Get started with AI-powered information extraction in Azure·Hosted VM in class · Azure for self-paced·Implement with Foundry (55–60%)·Episode 10

    • Content Understanding
    • Documents
    • Audio and video

Also on Microsoft Learn