Personal training advice

Live online training for international teams.For businesses
Find my course
AI Agents

AI-500 Designing and Implementing Multi-Agent AI Solutions: multi-agent systems on Azure

Design multi-agent systems in Microsoft Foundry, connect agents through A2A and MCP, measure their quality and take them into production securely.

After the course you can design architectures in which several AI agents work together: you break goals into workflows, agents and tools, choose orchestration patterns such as hub-and-spoke or an orchestrator with subagents, and connect agents through A2A and MCP. You plan memory and context, assess agents with tracing and evaluations in Microsoft Foundry, and apply guardrails, security and governance rules for production use. The course suits AI engineers, developers and solution architects who already run agentic systems in production, write Python confidently and want to prepare for the AI-500 exam.

AdvancedDE / ENBusiness training · Live online

What you’ll take away.

  • Design multi-agent architectures with suitable orchestration patterns
  • Build agents in Microsoft Foundry with Microsoft Agent Framework, MCP tools and memory
  • Plan agent-to-agent communication over A2A with shared state
  • Measure and improve agent systems with tracing and evaluations
  • Put security, governance and deployment in place for agent solutions
Microsoft · AI-500

Your training in detail

Who is it for?
AI engineers, developers and solution architects who plan and own production-ready agent systems on Azure.
Prerequisites
Experience with AI and machine learning development, with agentic systems in production and with orchestrating agents in Microsoft Foundry, confident Python, and experience with Azure compute, network, storage and data services.
Format & planning
Standard length: 4 training days. A different length can be arranged for companies.

Agenda: from understanding to application

Our independently developed agenda proposal. Focus areas, lab environment and scope are agreed before booking.

  1. Architecture: breaking goals into workflows, agents and tools, control loops and human-in-the-loop

  2. Orchestration: sequential, parallel, peer-to-peer, hub-and-spoke and orchestrator with subagents

  3. Prompt engineering, context management and RAG across multiple agents

  4. Tools and communication: MCP servers and clients, A2A, and short- and long-term memory with shared state

  5. Evaluate and monitor: tracing in Foundry, evaluations, cost and latency

  6. Secure and deploy: guardrails, identities, governance and deployment on Azure

PRACTICE & TRANSFER

You build a team of agents for a multi-step research task: an orchestrator agent splits the work, specialised subagents call tools through MCP, and you review the runs with tracing before adding guardrails.

Exams, licences & credentials

The course prepares you for exam AI-500, which counts towards Microsoft Certified: Multi-Agent AI Solutions Expert. The exam and its fee are included only if the offer confirms it. The certification is issued by Microsoft. This is not an official Microsoft course unless the offer states otherwise.

Vendor references & review date

Sources checked on 2026-11-10. These sources describe the technology or external exam requirements. They do not imply a partnership with NextNowa.

Your learning content

These objectives describe the training topic. The specific syllabus, dates and trainer are agreed before booking; the written offer is binding.

Individual offer: Availability and delivery are confirmed on enquiry.

Frequently asked questions

What experience should I bring?

Microsoft expects experience with AI development, with agentic systems in production and with orchestration in Microsoft Foundry, plus confident Python. If you have only built single agents so far, AI-103 is a sensible first step.

Which areas does the AI-500 exam cover?

Four areas: architecture, development on Azure, evaluation and monitoring, and security, governance and deployment. Building multi-agent solutions carries the largest weight.

Is it only about Microsoft tools?

The focus is Microsoft Foundry and Microsoft Agent Framework. Open standards such as MCP and A2A and frameworks such as LangGraph are covered as far as they matter for architecture decisions.

Does AI-500 Designing and Implementing Multi-Agent AI Solutions: multi-agent systems on Azure include a certificate?

The course prepares you for exam AI-500, which counts towards Microsoft Certified: Multi-Agent AI Solutions Expert. The exam and its fee are included only if the offer confirms it. The certification is issued by Microsoft. This is not an official Microsoft course unless the offer states otherwise.

KEEP LEARNING

Related topics in this field

View all topics

Also relevant

Guides

NextNowa Guide

Your training guide

What would you like to learn next?

Ask about courses, learning paths or advice. I will help you find your next step.

Automated guidance using reviewed information. No booking or funding confirmation. Please avoid sensitive data.