AI Engineering
Background summary — AI-generated; not source-grounded.
About
AI Engineering (O’Reilly, 2024) by Chip Huyen is a comprehensive guide to building production applications on top of pre-trained foundation models. The book defines AI Engineering as a distinct discipline from ML Engineering — covering the full application stack: prompt engineering, retrieval-augmented generation, evaluation design, fine-tuning, and deployment. It draws on Chip’s experience building AI products and consulting with enterprises across industries.
The central argument is that most AI product failures are failures of data quality, evaluation rigour, and user research — not model capability or infrastructure selection. On publication it became the most-read O’Reilly title since launch.
In the wiki
Mentioned by: Chip Huyen Episode: Chip Huyen on AI Engineering (Lenny’s Podcast, 2025)
Chip describes the book as defining the AI Engineering discipline. The viral LinkedIn table — contrasting what teams think improves AI apps (model selection, AI news, new frameworks) with what actually does (user conversations, data preparation, prompt engineering) — originated in the book and was Chip’s most-engaged post.
See also
- AI Engineering — concept page for the discipline
- Chip Huyen — author
- Designing Machine Learning Systems — Chip Huyen’s earlier book
- Evals
- Reinforcement Learning from Human Feedback