Large language models have evolved from passive text generators into active, tool‑using agents capable of reasoning, taking actions, and orchestrating complex workflows. With the rise of the Model Context Protocol (MCP), standardized function calling, and agent‑centric application patterns, developers are now building systems where LLMs collaborate with APIs, data sources, and business logic in real time. The question is no longer “How do I prompt a model?” but “How do I design an ecosystem where models can safely and reliably get things done?”

This workshop dives deep into modern agent development using Microsoft Foundry, OpenAI, and third‑party providers that support MCP and advanced function‑calling interfaces.

You’ll learn how today’s agentic systems work, how they reason, how they choose tools, and how to design the surrounding infrastructure that keeps them safe, predictable, and aligned with business requirements.

Key questions we’ll explore include:
• How do agent architectures differ from traditional LLM applications?
• What is the Model Context Protocol, and how does it enable interoperable, pluggable AI ecosystems?
• How do function‑calling models reason about which tools to use — and how do you design those tools effectively?
• How can agents integrate with enterprise data, APIs, and workflows while maintaining safety, governance, and auditability?
Through hands‑on exercises, you’ll build:
• Function‑calling interfaces that expose business capabilities to LLMs
• MCP servers and clients that allow models to interact with tools, data sources, and services
• Retrieval‑augmented agents that ground their decisions in organizational knowledge
• Agent workflows that chain reasoning, tool use, and verification steps
• Safety and evaluation layers that ensure agents behave reliably in production

By the end of the workshop, you’ll understand how to design, build, and operate agentic AI systems that go far beyond chat — systems that take action, automate processes, and integrate deeply with your applications and data.

Who should attend?
This workshop is ideal for developers and data scientists looking to deepen their understanding of generative AI and LLMs and integrate them into their projects.
Programming experience in C# or Python will be required for most of the hands-on labs.

What should you bring?
• Laptop with a development environment suitable for Python or C# development.
• Access to Microsoft Azure and Microsoft Foundry or the OpenAI API services.