Hands-On Microsoft Foundry
Published by O'Reilly Media, Inc.
Build, scale, and operationalize AI agents in Azure
What you’ll learn and how you can apply it
- Understand the architecture, capabilities, and agent lifecycle in Microsoft Foundry
- Learn how the Model Context Protocol (MCP) enables structured orchestration and tool invocation
- Discover how to use tools, memory, and goals to create intelligent agents
- Use best practices for deploying and monitoring agents in enterprise environments
- Build and simulate goal-based agents using Microsoft Foundry Studio
- Integrate MCP-wrapped tools and external APIs securely
- Design multi-agent workflows that collaborate across tasks
- Evaluate and deploy agents using trace inspection and telemetry
Course description
As generative AI and autonomous agents continue to transform enterprise workflows, organizations are increasingly turning to Microsoft Foundry—Microsoft’s unified, model-agnostic AI platform—to design, build, and scale secure orchestrated agent-based applications. Microsoft cybersecurity and AI leader Razi Rais demonstrates the practical tools and end-to-end workflows needed to plan, develop, and operationalize intelligent AI agents that can reason, remember, and take autonomous action, all while integrating into existing enterprise ecosystems.
Through a combination of interactive design principles and over 90 minutes of hands-on development demos, you’ll learn to build intelligent agents from the ground up—defining agent behaviors, connecting external and internal tools, managing persistent memory, and simulating task completion scenarios. You’ll gain a deep, working knowledge of Microsoft Foundry’s architecture and the Model Context Protocol (MCP) for structured tool and API integration. The course also covers best practices for debugging, refining, and scaling agent applications, using multi-agent workflows, trace inspection techniques, and enterprise-ready deployment pipelines.
This live event is for you because...
- You’re an architect, developer, or AI SME who’s building intelligent applications.
- You want to operationalize AI agents with a secure, governed Azure platform.
- You’re exploring how to scale beyond copilots into task-oriented agents.
Prerequisites
- (Optional) A free Microsoft account and either a Microsoft Azure 30-day trial or a paid Azure subscription (to follow along with demonstrations)
- Familiarity with cloud-based AI services and LLMs
- Basic experience with REST APIs and Azure services
- Development experience in Python or C#
Recommended follow-up:
- Read Building Applications with AI Agents (book)
- Take Zero Trust for AI Systems (live online course with Razi Rais)
Schedule
The time frames are only estimates and may vary according to how the class is progressing.
Introduction to agent development in Microsoft Foundry (55 minutes)
- Presentation: Core concepts—goals, memory, tools, and orchestrator; overview of Microsoft Foundry and Agent Service; agent lifecycle; how agents differ from chatbots or copilots
- Demonstration: Spinning up a basic goal-based agent using Microsoft Foundry Studio; simulating interaction and inspecting the goal trace
- Group discussion: What enterprise use case would you automate with an agent?
- Q&A
- Break
Building and operationalizing an agent (55 minutes)
- Presentation: Authoring goals and planner configurations; configuring memory; deploying agents and exposing them via REST APIs; using simulation output and logs
- Demonstration: Building and testing a memory-enabled agent; evaluating behavior using the simulation panel
- Q&A
- Break
Integrating MCP tools and APIs with your agent (55 minutes)
- Presentation: What is the Model Context Protocol?; roles of the MCP server and client; wrapping async tools as OpenAPI-compatible functions; registering and invoking tools securely; governance and authentication
- Demonstration: Registering an MCP-wrapped API tool; simulating tool invocation in an agent and reviewing the trace
- Group discussion: Which tools or APIs would you integrate first?
- Q&A
- Break
Designing and deploying multi-agent solutions (55 minutes)
- Presentation: Designing for modular workflows; defining agent roles (planner versus tool agents); conversation state and chaining; multi-agent topologies and deployment patterns
- Demonstration: Designing and simulating a planner agent coordinating with a tool agent
- Q&A
- Break
Future directions (20 minutes)
- Group discussion: The future of agents on Azure; integrating with Azure OpenAI; security best practices; and DevOps pipelines
- Q&A
Your Instructor
Razi Rais
Razi Rais is a cybersecurity and AI leader at Microsoft, with over 20 years of experience building secure, resilient systems for Fortune 500 companies worldwide. He brings a globally informed perspective to enterprise-scale security, shaped by hands-on experience living and working in Singapore, UAE, France, and the United States. His background spans engineering, architecture, and product management, giving him a unique lens on the evolving intersection of AI and cybersecurity. He drives enterprise efforts to secure AI at scale and address emerging threats using frameworks like the NIST AI Risk Management Framework, OWASP Top 10 for LLMs, and MITRE ATLAS. Razi has coauthored several books, including Zero Trust Networks, second edition, and Azure Confidential Computing and Zero Trust and is a frequent speaker at conferences such as RSA and Identiverse. He serves on the GIAC advisory board, is a Microsoft Certified Trainer and delivers AI and cybersecurity training worldwide. Connect with him on LinkedIn to follow his latest work.