Microsoft Semantic Kernel in Action

Build AI applications that integrate multiple models, data sources, and other components using Microsoft Agent Framework.

Even simple AI applications need to combine LLMs and other models, functions and plugins created in various programming languages, and custom data sources. With the Microsoft Agent Framework open-source development kit, you can create chatbots, copilots, and agents by chaining individual components together using a few lines of glue code. Agent Framework makes it a snap to build AI apps from scratch or integrate AI features into an existing codebase.

In Building AI Agents in .NET you'll learn how to:

*; Implement short-term and long-term memory in AI applications
*; Create and manage plugins
*; Design and implement multi-modal AI applications
*; Orchestrate complex AI tasks using function calling and planners

Building AI Agents in .NET guides you from basics like creating and connecting your first plugins to advanced features including Agent Framework's powerful AI-powered orchestration capabilities. Everything you learn is put into practice with hands-on projects such as a ';Smart Garden' AI assistant and a ';Rain Detector' assistant that analyzes temperature, humidity, and luminosity, and then creates analysis charts in real time using generative AI. Youll learn AI orchestration techniques to achieve autonomous and semi-autonomous systems (human-in-the-loop) using tools and planners, how to use open-source small language models for on-premises systems, advanced techniques for decomposing prompts into native and semantic plugins, and much more. Plus, the book's focus on enterprise-ready solutions ensures you're always considering security, scalability, and responsible AI practices.

About the book

Building AI Agents in .NET teaches you to create AI applications and integrate AI into existing software using C# and Microsoft Agent Framework. The book covers both proprietary AI models like OpenAIs GPT and open-source options like Llama and Phi. As you work hands-on through several interesting and relevant projects, youll explore assorted connectors and AI services, orchestrate complex tasks, and implement long-term memory using RAG.

About the reader

For beginner and expert .NET developers, data scientists, and AI enthusiasts seeking to integrate AI into applications.

About the author

Daniel Costea is a veteran software engineer with over two decades of experience in enterprise solutions. A Microsoft MVP since 2019, he specializes in .NET technologies, IoT, machine learning, and generative AI. Daniel is a frequent speaker at technology events, has trained hundreds of students, and developed multiple courses on cutting-edge topics.

Get a free eBook (PDF or ePub) from Manning as well as access to the online liveBook format (and its AI assistant that will answer your questions in any language) when you purchase the print book.