Implementing Agents and Copilots using Semantic Kernel and Azure AI Agent Service (AZAGENTS)

Course Overview

This intensive five-day online class will transform how you build intelligent AI solutions using Microsoft's Semantic Kernel framework and Azure AI services. Beginning with Azure AI Foundry fundamentals, you'll master essential concepts like prompt engineering techniques and Retrieval Augmented Generation (RAG) for integrating enterprise data with LLMs, as well as integrating AI skills using Model Context Protocol (MCP) and Agent2Agent (A2A) protocols. You'll also learn how to deploy AI applications using the Azure Developer CLI, as well as ensuring a predictable outcome using Evaluations.

You'll then dive deep into Semantic Kernel development, exploring its core components, chat history integration, and multi-modal capabilities while learning to optimize prompts using various template formats. The course progresses to advanced plugin development, where you'll implement native functions, integrate existing APIs via OpenAPI, and utilize planners with function calling capabilities. In the more advanced sections, you'll work with Kernel Memory for vector search, implement sophisticated multi-agent solutions with personas, and orchestrate complex workflows using the Semantic Kernel Process Framework with Orleans and Dapr runtimes.

In Develop Agents using Azure AI Agent Service, you'll master Azure AI Agent Service to build sophisticated AI agents capable of taking complex actions and accessing knowledge sources. You'll implement and integrate diverse action tools including Code Interpreter, Function Calling, and OpenAPI tools, while also leveraging knowledge tools such as File Search, Azure AI Search, and Bing Grounding to enhance your agents' capabilities. The module culminates with practical techniques for orchestrating these agents using Semantic Kernel framework and implementing essential production safeguards through content filtering and comprehensive observability practices.

Building on the foundational knowledge of Semantic Kernel and Azure AI Agent Service from previous modules, this module focuses specifically on integrating these technologies within the Microsoft 365 ecosystem. You'll learn how to extend Microsoft 365 Copilot with custom plugins and connectors, understand the differences between declarative and custom engine agents, and implement practical Semantic Kernel-based solutions that seamlessly integrate with Microsoft's productivity suite. Throughout all modules, you'll work with production-grade code samples in both Python and C#, providing you with immediately applicable skills for building enterprise-ready AI agent solutions.