Microsoft Agent Framework: Orchestrating AI LLMs

August 20, 2026 | Online

Mike Yeager

The true power of Large Language Models (LLMs) isn’t found in a chat box, but in their ability to act as the "reasoning engine" within a complex application. This intensive one-day workshop moves beyond simple API calls to explore the architecture of AI Agents. Participants will learn how to build systems that can self-correct, use external tools (browsers, databases, APIs), and maintain state across multi-step reasoning chains. We will focus on the practical orchestration of LLMs within professional codebases to create reliable, production-ready agentic workflows.

This is an online event.

Click here to register for this event!
Sign up today and save $ 150!  Only $ 850!
(Early bird price is only good until 07/20/2026)

AGENDA

AGENDA

Workshop Modules

• Module 1: From chats to agents – Beyond text and image generation and rudimentary semantic lookups. Understanding the shift from simple, generic chat bots to complex, autonomous reasoning.
• Module 2: Tool use & agent knowledge – Teaching LLMs how to interact with your existing APIs, documents, and databases securely. We’ll cover custom functions as well as Model Context Protocol (MCP), semantic databases, structured output, and more.
• Module 3: Memory & state management – Strategies for persistent context and "long-term memory" in agentic systems as well as how to handling long running processes.
• Module 4: Multi-agent aystems & workflows – Orchestrating specialized agents (e.g., a "coder" agent and a "reviewer" agent) to solve complex tasks.
• Module 5: Evaluation & reliability – How to test, monitor, and debug non-deterministic agentic behavior. OpenTelemetry, observability, and middleware.

Hands-On Lab

• Hands-On Lab: Architecting a "research & action" agent – Build an agent that fetches real-time data and updates a database autonomously.