Microsoft Foundry: Build, Orchestrate, and Automate AI Agent

Understand what AI agents are and explore the core elements that enable them to function intelligently and autonomously.

See how prompt-based interactions evolved into dynamic, decision-making agents.

Explore the core traits that give AI agents the ability to act independently and make decisions without constant human input.

Learn why clear, actionable instructions are essential for agents to function effectively, beyond just having access to data.

Discover how agents operate with both fixed and evolving goals—and what that means for their adaptability and performance.

Task-based agents, conversational agents, data-driven agents, and workflow agents.

Foundry’s role in Microsoft’s AI ecosystem and how it differs from other AI platforms.

No-code approach vs custom development: benefits and limitations.

Models, agents, tools, data sources, workflows, and governance.

Conceptual explanation of visual builders, configuration panels, and prompts.

Step-by-step: create an agent that answers a set of FAQ questions. Configure the model and prompt using the UI, then save the agent.

In this lesson, we'll cover testing and iterating your agent.

In this lesson, we'll look through the templates that Microsoft Foundry provides.

In this lesson, we'll give you ready-to-go instructions to create your first agent.

Step through configuring your workspace to start building and running Foundry agents effectively.

Discover the various tools available in Foundry and how they enhance agent capabilities.

Understand the types of models Foundry provides and how they can be applied in different agent scenarios.

Learn to create workflows that orchestrate agents, tools, and models for practical tasks.

Explore best practices for ensuring ethical, safe, and responsible AI use within Foundry.

Understand the difference between a single agent and a workflow of agents. Learn when to use multi-agent orchestration for complex tasks (e.g. separate agents for research, strategy, and execution).

Step-by-step: create a sequential workflow where Agent A’s output feeds into Agent B, then Agent C. For example, Agent 1 gathers data, Agent 2 summarizes it, Agent 3 generates a final report.

Learn how to add approval steps: the workflow pauses to ask the user a question or confirmation before proceeding (useful for quality checks or manual overrides).

Overview of advanced flows: you can route to different agents based on conditions or let agents collaborate in a group chat format. (This is useful for fallback or expert routing scenarios.)

Explore example applications (non-industry-specific): e.g. a support chatbot, an internal knowledge bot, or a scheduling assistant. See how Foundry’s features fit these scenarios.

Build an agent designed to handle customer inquiries efficiently and effectively.

Create an agent that gathers, analyzes, and summarizes data for actionable insights.

Set up an agent to manage and provide access to organizational knowledge.

Develop an agent that automates repetitive tasks, improving productivity and consistency.

Design and implement your own multi-agent solution that combines prompts, tools, and workflows in a practical scenario.

Review the key concepts learned. Learn about further learning paths (advanced Foundry features, coding with the SDK, or Microsoft Copilot Studio) and how to stay current with AI tools.

In this lesson, you will learn how to leave a review for the course.

In this lesson, you will learn how to get a certificate.

You’ve completed the course—now it’s time to go beyond! In this bonus lecture, you’ll discover how to:

✅ Get support in your career growth with exclusive communities and resources

✅ Develop new knowledge and stay ahead with advanced AI skills

✅ Enhance your expertise with additional courses and hands-on challenges

Don’t stop here—keep growing, keep learning, and unlock new opportunities!