Final review
AI-200 Exam Preparation
Exam format, the official skills measured, a service decision matrix, CLI reference, and the details that most commonly cost marks. Structured for review in the final week before your exam.
Exam format and scoring
Study priority by domain weight
Allocate revision time in proportion to exam weight. Domain 2 carries the most marks and contains the material least likely to be covered by prior AZ-204 experience.
Module completion
Modules you have marked complete are highlighted. Progress is stored locally in your browser.
Frequently asked questions
What score do I need to pass the AI-200 exam?
You need 700 out of 1000. The score is scaled, not a simple percentage, so it does not map directly to a number of correct answers.
How many questions are on the AI-200 exam?
Between 40 and 60 questions in 100 minutes. Question types include single-answer multiple choice, multiple response, drag-and-drop ordering, and case studies with a shared scenario.
Which AI-200 domain carries the most weight?
Domain 2, Develop AI solutions using Azure data management services, at 25–30%. This covers Cosmos DB vector search, PostgreSQL with pgvector, and Azure Managed Redis.
How should I revise the night before the AI-200 exam?
Focus on the decision matrix and the common gotchas rather than deep technical detail. Most exam questions ask you to select the correct service for a scenario, and the gotchas are the details that separate two plausible-looking answers.
Is the AI-200 certification worth it?
The AI-200 replaces AZ-204 as the Microsoft associate-level developer certification for Azure. It is currently the primary credential for developers building AI-powered applications on Azure infrastructure.