AI Engineering: Building Applications with Foundation Models
by Chip Huyen (O'Reilly)
The definitive practitioner's guide to building production applications on top of foundation models.
Overview
AI Engineering: Building Applications with Foundation Models by Chip Huyen (O'Reilly, 2025) lays out a practical framework for developing and deploying applications on top of foundation models, and explains how AI engineering differs from traditional ML engineering and what the emerging AI stack looks like. The book walks through navigating the AI ecosystem — models, datasets, and evaluation benchmarks — and covers the core techniques a practitioner needs: model evaluation and selection, prompt engineering, retrieval-augmented generation (RAG) and context construction, finetuning versus prompting trade-offs, dataset engineering, inference optimization, and designing end-to-end AI application architectures with feedback loops. It became the most-read book on the O'Reilly platform in 2025 and has been translated into more than a dozen languages, reflecting its adoption as a standard reference for teams building with LLMs. Supporting materials and resources are maintained by the author in a companion repository.
At a Glance
- Topic
- Models
- Level
- Intermediate
- Format
- Book
- Cost
- Paid
- Duration
- ~500-page book, self-paced
- Provider
- Chip Huyen (O'Reilly)
- Hands-on
- No
- Certificate
- None
What You’ll Learn
- ✓How AI engineering differs from traditional ML and what the modern AI application stack looks like
- ✓How to evaluate and select foundation models, including building evaluation pipelines and benchmarks
- ✓Prompt engineering, retrieval-augmented generation (RAG), and context construction for reliable outputs
- ✓When to finetune versus prompt, plus dataset engineering for adapting foundation models
- ✓Inference optimization and designing end-to-end AI application architectures with feedback loops
Highlights
- •By Chip Huyen, author of 'Designing Machine Learning Systems'
- •Most-read book on O'Reilly in 2025; translated into 13+ languages
- •Model-agnostic, production-focused framework rather than framework-specific tutorials
- •Companion resources maintained in the author's open aie-book repository
Who It’s For
Best For
- ✓Software engineers moving into building LLM/foundation-model applications
- ✓ML practitioners transitioning from model training to application engineering
- ✓Tech leads designing production AI systems and evaluation strategy
Prerequisites
- •General software engineering experience
- •Basic familiarity with machine learning and LLM concepts (helpful but not strictly required)
FAQ
What is AI Engineering: Building Applications with Foundation Models?
A comprehensive O'Reilly book (2025) by Chip Huyen on the discipline of AI engineering — building real applications with readily available foundation models. Written for software engineers and ML practitioners moving from demos to reliable, production-grade LLM products.
Is AI Engineering: Building Applications with Foundation Models free?
AI Engineering: Building Applications with Foundation Models is a paid resource.
What level is AI Engineering: Building Applications with Foundation Models for?
AI Engineering: Building Applications with Foundation Models is aimed at a intermediate audience. Recommended background: General software engineering experience, Basic familiarity with machine learning and LLM concepts (helpful but not strictly required).
How long does AI Engineering: Building Applications with Foundation Models take?
Expect roughly ~500-page book, self-paced. Most learners work through it at their own pace.
What will I learn from AI Engineering: Building Applications with Foundation Models?
You'll learn: How AI engineering differs from traditional ML and what the modern AI application stack looks like; How to evaluate and select foundation models, including building evaluation pipelines and benchmarks; Prompt engineering, retrieval-augmented generation (RAG), and context construction for reliable outputs; When to finetune versus prompt, plus dataset engineering for adapting foundation models; Inference optimization and designing end-to-end AI application architectures with feedback loops.