AI-500 Multi-Agent AI Solutions Expert in German

Who it is for

For .NET developers and architects who have to take multi-agent systems into production or who are aiming at exam AI-500 and already know AI-103.

Prerequisites

The AI-103 course (at least modules I1 and I2) and the AI track (at least T6 to T10) or equivalent experience, plus K3 (Bicep, managed identity, CI/CD) and K4 (decision records). You need an Azure account only for the optional steps. The exam expects Python experience and names LangGraph: the course teaches .NET and only reads Python.

Afterward, you can

• Break a goal into workflows, agents, tools, personas and autonomy levels and design protocols, memory and state for it
• Build agent teams with the Microsoft Agent Framework: orchestration patterns, MCP servers, approvals, caching and scaling
• Shape prompts, context, memory and knowledge for several agents
• Evaluate and monitor agent teams and steer their usage and cost
• Design identity, secrets, guardrails and tests for an agent system
• Ship an agent system across environments with canary, blue/green and rollback

Contents

1. X0

   Orientation

   map of the exam, AI-103 and the course, the exercise account for multi-agent runs

   2 h
2. X1

   Architecture

   decomposing goals, protocols, memory and state, technology components, zero trust for agents

   8 h
3. X2

   Develop

   prompts, context, memory, multi-agent RAG, tools, MCP, orchestration, approvals, caching, A2A, other frameworks

   19 h
4. X3

   Evaluate, optimize, monitor

   evaluation, continuous improvement, duration and context problems, tracing, tokens, cost and availability

   8 h
5. X4

   Secure, govern, deploy

   access and sign-in flows, Key Vault, red teaming, guardrails, release and rollback, tests and CI/CD

   11 h
6. X5

   Wrap-up

   final project: the group-travel team as a design and sketch, defending the design, checking gaps

   4 h

About 52 hours in all (indicative).

Lessons

You see the text and video of every lesson here on the site after buying the course; the titles and goals are already shown. A self-study course: you work through the lessons in order and check your exercises yourself against the solutions.

You can read the first module (X0) for free, with text and video, without signing in: so you see how the course works before you buy.

The lessons stand under their modules, in the order of the course. Click a module to open or close its lessons.

Final project: the Kladde

Certifications

An offer, never a requirement: these Microsoft exams fit this stage. You take them at Microsoft; we are not a Microsoft training partner.

• AI-500 Multi-Agent AI Solutions Expert

  Requires AI-103 and covers Agent Framework and MCP; expects Python skills.

The course follows the topics Microsoft lists in the study guide of exam AI-500 (the list carries no date; read on Microsoft Learn on November 12, 2026) and replaces neither Microsoft's material nor exam preparation. Whether the exam is already regularly offered is open: check its status with Microsoft. The exam is offered in English only and expects Python experience; the course teaches .NET. You take the exam with Microsoft; registration, price and dates are there.

Keep learning with Microsoft Learn

Microsoft's learning paths and documentation that go with this course.

• Study guide for Exam AI-500: Designing and Implementing Multi-Agent AI Solutions
• Microsoft Certified: Multi-Agent AI Solutions Expert
• Microsoft Agent Framework
• Agents in Microsoft Foundry