Hands-On Large Language Models: Building and Using LLMs from Scratch with Python
What if you could stop reading about AI and start building it?
Large language models have quietly become the most important technology of our generation - powering everything from customer service to code generation to scientific discovery. But most books on the subject leave you stuck in one of two places: drowning in academic theory with no code in sight, or copy-pasting API calls with no real understanding of what's happening underneath.
Hands-On Large Language Models is different. It's the guide the author wishes existed when they started: a complete, practical path from "What actually is a transformer?" to building, fine-tuning, evaluating, and deploying real LLM-powered applications - with working Python code in every single chapter.
Whether you're a developer pivoting into AI, a student trying to get ahead of the curve, or an engineer who wants to move beyond prompt-and-pray development, this book gives you both the why and the how.
Inside, you'll learn how to:
Understand exactly how transformers, tokens, embeddings, and attention actually work - no hand-waving, no PhD requiredSet up a professional LLM development environment with proper credential security and cost controlsMaster prompt engineering as a real, testable engineering discipline - not guessworkBuild production-grade applications with conversation management, streaming, and error handlingBuild a Retrieval-Augmented Generation (RAG) pipeline from scratch, including chunking, embeddings, and vector databasesFine-tune open-source models using LoRA and QLoRA - even on modest hardwareEvaluate LLM outputs systematically, catch hallucinations, and build regression tests that actually workBuild autonomous AI agents that reason, use tools, and complete multi-step tasks safelyDeploy, monitor, and scale your LLM applications in production - including defending against prompt injectionEvery chapter combines deep conceptual grounding with full, runnable code examples - so you're never just reading about a concept, you're building it, breaking it, and understanding exactly why it works.
By the end of this book, large language models will stop feeling like a mysterious black box and start feeling like exactly what they are: a powerful, understandable tool you can bend to your own ideas.
This book is for you if:
You know some Python and want a serious, structured path into LLM engineering
You're tired of tutorials that skip the "why" and just paste API keys
You want to build real, deployable applications - not just toy demos
You want to understand LLMs deeply enough to debug them, not just use them
Stop reading about AI from the sidelines. Open your editor, and start building.
Customer Reviews
ThriftBooks ® and the ThriftBooks ® logo are registered trademarks of Thrift Books Global, LLC
