AB-100 study guide
AB-100 · Work through your exam, one topic at a time.
AB-100 is the required exam for the Microsoft Certified: Agentic AI Business Solutions Architect Expert credential. It targets accomplished solution architects who plan, design, and deploy AI-driven business solutions across Microsoft platforms - Dynamics 365, Power Platform, Copilot Studio, Azure AI services and Azure OpenAI - and who lead an agentic-first architecture strategy rather than configure individual features. The exam measures three areas: planning AI-powered business solutions (35–40%), designing AI-powered business solutions (30–35%), and deploying them (20–25%). Expect emphasis on multi-agent orchestration, securing models and data workflows, open protocols such as Agent2Agent and Model Context Protocol, lifecycle and environment strategy, and return-on-investment analysis. Topics on this exam: Plan AI-powered business solutions, Design AI-powered business solutions, and Deploy AI-powered business solutions. The AB-100 study guide has 3 sections and about 1,220 words.
01Plan AI-Powered Business Solutions
Planning questions test whether you can decide whether and where an agent belongs before anything is designed. Expect scenarios that hand you a process description and ask for the correct next step, the right success metric, or the right target workload. Key concepts: Value assessment : baseline the current process (cycle time, cost per transaction, containment/deflection rate, error rate) before defining targets. A plan without a baseline cannot prove value later.
421 words, about 2 min02Design AI-Powered Business Solutions
Design answers usually fail because the candidate picks a component before deciding which layer the requirement belongs to. Separate the solution into layers: Experience : where the agent surfaces - Microsoft 365 Copilot, Teams, a website, a custom app, or voice. Orchestration : the agent itself - instructions (persona, tone, scope, guardrails), topics for scripted conversation paths, and generative orchestration that selects knowledge and tools dynamically. Knowledge : Dataverse tables, SharePoint, uploaded documents, websites, connectors.
402 words, about 2 min03Deploy AI-Powered Business Solutions
Deployment questions test ALM discipline and operational readiness. Environment strategy : development → test → production, each with its own Dataverse instance. Author as unmanaged; ship managed. Solution hygiene : no hardcoded values - use environment variables; no embedded credentials - use connection references so each environment supplies its own connection.
397 words, about 2 min