An up-to-date and hands-on tutorial for building production-grade LLM applications In The AI Engineering Bootcamp: Build, Ship, Share, AI Makerspace co-founders “Dr. Greg” Loughnane and Chris “The Wiz” Alexiuk guide the reader through the foundational concepts and code needed to build production-grade Large Language Model (LLM) applications using leading open-source tooling. The book explains the four primary design patterns of generative AI-prompt engineering, Retrieval Augmented Generation (RAG), fine-tuning, and agentic reasoning-and how to leverage them as first principles to build scalable LLM applications that are high-performance and efficient. You’ll find classroom-tested lessons that offer immediate insights into building LLM applications, as well as example projects to give you hands-on experience that can be immediately applied within our company or with your clients today. The AI Engineering Bootcamp provides everything you need, from your initial AI-assisted Interactive Development Environment (IDE) set up and first deployment to the boilerplate Python code you need to prototype LLM, RAG, agent, and multi-agent applications. The book also covers how to prepare your prototypes for production by setting up open-source LLM and embedding model endpoints, leveraging caching for prompts and embeddings, what you need to host and deploy your application on premise, and more! Of course, the book also includes a discussion of how to leverage emerging protocols, including Model Context Protocol (MCP) and Agent2Agent protocol, which have taken the agent landscape by storm in 2025. While the authors provide an opinionated view of the best-practice open-source tooling based on the current landscape, throughout this book, you will avoid vendor lock-in to any specific Cloud Service Provider (e.g., Amazon Web Services, Google Cloud Platform, or Microsoft Azure).
