Practical AI Agents with Microsoft Agent Framework: A hands-on guide for .NET developers

Build AI agents that don’t just chat—they take action.AI is transforming how software works, but most learning resources focus on Python. If you’re a .NET developer, you’ve likely struggled to find clear, practical guidance—until now.This book shows you how to build real AI agents using C# and the Microsoft Agent Framework. You’ll learn how to create systems that understand natural language, call functions, maintain memory, perform semantic search, and integrate with external services.This is not theory.This is production-oriented code you can run today.What you’ll learn and buildStarting from your first Hello Agent, you’ll progressively master:• Tools and function calling to let agents take real actions• Short-term context and long-term semantic memory• Retrieval-Augmented Generation (RAG) with vector stores• Multi-agent systems and orchestration patterns• Native workflows: sequential, concurrent, and dynamic routing• MCP integration for external tool servers• Structured output for reliable, typed responsesThe book culminates in ExpenseTracker — a complete, production-ready application featuring natural language expense recording, receipt parsing with Vision AI, a Telegram bot, and a REST API.Why this book?• A progressive learning path where each chapter builds on the previous one• Every concept demonstrated with runnable, real-world code• A complete end-to-end project, not isolated examples• Built on the official Microsoft Agent Framework, not wrappers or hacks• Full source code available on GitHub, ready to clone and runWho this book is for• .NET developers who want to add AI capabilities to their applications• C# programmers exploring AI agents and LLM-based systems• Software architects designing intelligent, distributed solutions• Developers who prefer learning by building real softwarePrerequisites: solid C# knowledge and basic .NET experience.No prior AI or machine learning background required.About the authorMatteo D’Avena is a software architect and CTO with nearly 30 years of experience in software development. His career spans business intelligence in banking and finance, followed by over two decades building complex systems across multiple industries.,As CTO and co-founder of a company behind a leading SaaS platform for 3D configuration and AR visualization, he brings deep expertise in Microsoft .NET and enterprise architecture. In recent years, his focus has shifted to AI development, applying large language models and agent-based systems to real-world production products.This book reflects that hands-on experience: practical patterns that work in production, written for developers who build software for a living.Stop reading about AI. Start building with it. Read more