Large Language Models That Work: A Practical Python Guide to Transformers, Embeddings, Semantic Search, RAG, Fine-Tuning, AI Agents, Evaluation, and P

Large language models can generate impressive text in minutes. Turning that capability into a reliable product is a different problem. Retrieval can miss the decisive source. A prompt can break on the next edge case. An agent can call the wrong tool. A model upgrade can improve a benchmark while making your real workflow slower, costlier, or less accurate.

Large Language Models That Work is a practical, production-first guide to designing AI applications that are bounded, measurable, and maintainable. Using clear mental models, original diagrams, and focused Python examples, Liam Everly shows how transformers, embeddings, semantic search, retrieval-augmented generation, fine-tuning, tools, and agents fit into one dependable system.

Inside, you will learn how to: