AI-901 cheat sheet

Exam at a glance

Domain weightings

How much of the exam each domain covers. Spend your study time in proportion — the heavier the domain, the more questions you'll see.

Key terms

Microsoft Foundry

Microsoft Foundry is Microsoft's unified platform for building, deploying, and managing generative AI apps and agents on Azure. AI-901 requires implementing lightweight AI solutions in the Foundry portal and with the Foundry SDK.

Foundry SDK

The Foundry SDK is the client library used to build applications that call models and agents deployed in Microsoft Foundry. AI-901 covers building a lightweight chat client and agent client with it.

Generative AI

Generative AI is a class of AI that creates new content — text, images, code, or audio — from patterns learned during training. AI-901 covers how generative AI models work and their common business scenarios.

Agent

An agent is an AI solution that uses a model plus instructions, tools, and knowledge to accomplish tasks rather than answer a single prompt. AI-901 covers creating and testing a single-agent solution in the Foundry portal.

Prompt

A prompt is the system or user instruction given to a generative model to shape its response. AI-901 requires creating effective system and user prompts for generative AI models.

Responsible AI

Responsible AI is Microsoft's framework of principles — fairness, reliability and safety, privacy and security, inclusiveness, transparency, and accountability — for building trustworthy AI. AI-901 requires describing each principle.

Multimodal model

A multimodal model is a model that accepts and reasons over more than one input type, such as text plus images or audio. AI-901 covers interpreting visual input and responding to spoken prompts with multimodal models.

Computer vision

Computer vision is the AI field concerned with interpreting images and video — classification, object detection, and OCR — which is distinct from image generation. AI-901 covers computer vision alongside image-generation models and building a lightweight vision app.

Text analysis

Text analysis is the set of techniques for extracting meaning from text, including keyword extraction, entity detection, sentiment analysis, and summarization. AI-901 covers these techniques and building a lightweight text-analysis app.

Sentiment analysis

Sentiment analysis is a text-analysis technique that determines the emotional tone — positive, negative, or neutral — of a piece of text. AI-901 lists it among the common text-analysis techniques.

Entity detection

Entity detection is a text-analysis technique that identifies and categorizes named items such as people, places, and organizations in text. AI-901 lists it among the common text-analysis techniques.

Content Understanding

Azure Content Understanding in Foundry Tools is a service that extracts structured information from documents, images, audio, and video. AI-901 requires using it to build information-extraction solutions.

Information extraction

Information extraction is the AI workload that pulls structured data — fields, entities, tables — out of unstructured documents and media. AI-901 covers implementing it with Content Understanding.

Speech recognition

Speech recognition is the AI capability that converts spoken audio into text (speech-to-text); its counterpart, speech synthesis, converts text into spoken audio. AI-901 covers both and Azure Speech in Foundry Tools.