AI-901 Study Guide

A recommended approach to preparing for the Azure AI Fundamentals (AI-901) exam.

How to Use This Guide

The sidebar mirrors the two exam domains. Each topic page contains:

• Overview — what the concept is and why it matters for the exam
• Key Concepts — terms and definitions you must know
• Azure Services / Foundry Features — which tools are relevant
• Study Resources — links to Microsoft Learn modules and official docs

Recommended Study Order

1. Domain 1 — AI Concepts & Capabilities

   • Start with Responsible AI principles (always tested)
   • Learn how generative AI models work and how to choose between them
   • Study the common AI workload types and which Azure service handles each

2. Domain 2 — Implement with Microsoft Foundry (priority — 55–60% of exam)

   • Get hands-on with Azure AI Foundry — deploy a model, run the playground
   • Build the lightweight chat client from the Foundry SDK quickstart
   • Practice creating and testing a single-agent solution
   • Work through text, speech, vision, and information extraction scenarios

Suggested Study Timeline

Week	Focus
Week 1	Domain 1: Responsible AI, AI models, workload types
Week 2	Domain 2: Foundry portal — generative AI, agents
Week 3	Domain 2: Text/speech, computer vision, information extraction
Week 4	Full review + practice scenarios + exam sandbox

Hands-on Labs

Practical experience is essential for this exam — especially for Domain 2.

Resource	Notes
Azure AI Foundry portal	Free to explore; deploy models in the playground
Foundry SDK quickstarts	Python SDK examples for each AI capability
mslearn-ai-fundamentals labs	Official hands-on labs
mslearn-ai-studio labs	Foundry / AI Studio focused labs
AI-901T00 course	Free self-paced learning path

Python Knowledge You Need

You don't need to write Python code from scratch, but you must be able to read and understand it. Focus on:

# Pattern 1: Foundry SDK — chat client
from azure.ai.inference import ChatCompletionsClient
from azure.ai.inference.models import SystemMessage, UserMessage
from azure.core.credentials import AzureKeyCredential

client = ChatCompletionsClient(
    endpoint="<your-endpoint>",
    credential=AzureKeyCredential("<your-key>")
)

response = client.complete(
    model="<deployment-name>",
    messages=[
        SystemMessage("You are a helpful assistant."),
        UserMessage("What is Azure AI Foundry?"),
    ]
)
print(response.choices[0].message.content)

# Pattern 2: Azure AI Language — text analysis
from azure.ai.textanalytics import TextAnalyticsClient
from azure.core.credentials import AzureKeyCredential

client = TextAnalyticsClient(endpoint="<endpoint>", credential=AzureKeyCredential("<key>"))
documents = ["Azure AI Foundry is a unified platform for AI development."]
result = client.analyze_sentiment(documents)
print(result[0].sentiment)  # "positive", "negative", or "neutral"

Exam Day Tips

tip

• Domain 2 is 55–60% — spend most of your study time on Foundry hands-on work.
• Know the Foundry workflow: create a project → deploy a model → test in playground → build a client app.
• Understand Content Understanding — this is a newer service for extracting structured data from documents, images, audio, and video.
• Agents: know the difference between a single-agent solution (one model + tools) vs. a simple chat application.
• Responsible AI: the six principles are always tested — Fairness, Reliability & Safety, Privacy & Security, Inclusiveness, Transparency, Accountability.