The exam
AI-103 validates the skills of an Azure AI engineer who plans, builds, and operates generative AI and agentic solutions on Microsoft Foundry. Covers choosing and configuring Foundry models and services, building RAG and multi-agent applications, implementing responsible AI guardrails and observability, and implementing computer vision, text analysis, and information-extraction solutions including speech and document intelligence. It succeeds AI-102 (Azure AI Engineer Associate) as Microsoft's certification for engineers building on the Foundry platform.
Who it is for
As a candidate for this Microsoft Certification, you're an Azure AI engineer who builds, manages, and deploys agents and AI solutions that take advantage of Microsoft Foundry. For this exam, you should have experience developing apps by using Python, and you need to be familiar with the capabilities of general AI, generative AI, and Azure services. Your responsibilities include planning and managing Azure AI solutions, implementing generative AI and agentic solutions, implementing computer vision solutions, implementing text analysis solutions, and implementing information extraction solutions. In this role, you collaborate with business stakeholders, solution architects, data scientists, DevOps engineers, and cloud security engineers to design, implement, and maintain AI solutions.
What it covers
Your readiness map in the app is organised the same way, so what you see here is exactly what you study against.
What Preppio gives you
Across every domain above, in 5 formats: concept, scenario, ordering, multi-select, matching. Each one explains why the right answer is right and every distractor wrong.
Per domain, whether you are ready, weak, or not started — the same domains and weights as the exam itself.
Your study partner. He tracks where you are weak and drills exactly that, instead of serving you what you already know.
Try one
Answer on the page. The explanation appears when you submit; use Next to move through the formats.
Sample 1 · Concept
You are building a customer support application for Fabrikam, Inc. that must summarize incoming support tickets on a resource-constrained edge device with no internet connectivity. The solution must minimize compute requirements while still producing coherent summaries. What should you recommend?
Pick one, then submit.
Study links
FAQ
50 questions in 100 minutes.
70%, a scaled score of 700.
5 domains, weighted as the exam weights them: Plan and manage an Azure AI solution (28%), Implement generative AI and agentic solutions (33%), Implement computer vision solutions (13%), Implement text analysis solutions (13%) and Implement information extraction solutions (13%). Your readiness map in the app is organised by the same domains, so what you study against is what the exam tests.
Five formats: concept, scenario, ordering, multi-select and matching. The practice questions use every one of them, and the sample on this page shows one of each.
AI-103 is on the Single, Dual and All Access plans. You start on the free tier and add it from the plan picker inside the app. Three fundamentals exams (AZ-900, CLF-C02 and CDL) are always free, and the sample on this page is free to try.
Every question is generated from the exam's published objectives, then checked for accuracy, answer key and explanation before it is served. Nothing is copied from any exam, questions rotate by how often they have been seen, and they are retired when the objectives change.
