CertPrep Coach

AI-901 study guide and exam prep

The big picture first, then each skill area: what it covers, how the pieces fit together, and the traps the exam likes to set. Every guide ends with practice questions for that area.

AI-901 at a glance AI-901 has two halves. The first checks that you understand AI concepts: responsible AI, how models work, and the main AI workloads. The second checks that you can build those workloads in Microsoft Foundry: apps and agents, text and speech, vision, and information extraction. Identify AI concepts and capabilities (40–45%) Implement AI solutions with Microsoft Foundry (55–60%) Responsible AI six principles that apply to everything you build AI models and configuration how generative models work, choosing and deploying them AI workloads text, speech, vision, extraction, generative and agentic AI Text and speech solutions text analysis, Azure Speech, spoken prompts Generative AI apps and agents prompts, deployments, chat clients, agents in Foundry Vision and image generation images in prompts, generating images, vision apps Information extraction Content Understanding for documents, images, audio and video
AI-901 has two halves. The first checks that you understand AI concepts: responsible AI, how models work, and the main AI workloads. The second checks that you can build those workloads in Microsoft Foundry: apps and agents, text and speech, vision, and information extraction.

Follow the 2-week AI-901 study plan

Identify AI concepts and capabilities 40–45% of the exam

Responsible AI Microsoft's six responsible AI principles, what each one means in practice, and how to tell them apart in exam scenarios. AI models and configuration How generative models produce text, how to choose a model, and what deployment types and generation settings control. AI workloads Recognize the common AI workloads, including text analysis, speech, computer vision, information extraction, and generative and agentic AI, and match each to a scenario.

Implement AI solutions by using Microsoft Foundry 55–60% of the exam

Generative AI apps and agents Prompts, model deployments, chat clients built with the Foundry SDK, and single-agent solutions in Microsoft Foundry. Text and speech solutions Build apps that analyze text, use Azure Speech for speech to text and text to speech, and answer spoken prompts with a multimodal model. Vision and image generation Use multimodal models to interpret images, generate new images from prompts, and build lightweight apps with vision capabilities. Information extraction Extract content and structured fields from documents, images, audio and video with Azure Content Understanding in Foundry Tools.

Ready to test yourself?

Practice by skill area or take a timed mock exam. Every answer comes with an explanation.

Start practicing