Chip Huyen
Writer and computer scientist; author of AI Engineering
Background and ML tooling work
Chip Huyen describes herself as a writer and computer scientist who grew up in a small rice-farming village in Vietnam[1]. She graduated from Stanford[1]. She says her work is about bringing AI into production[1].
Huyen previously worked on machine learning tooling at NVIDIA, where she was a core developer of NeMo, and at Snorkel AI and Netflix[1]. She also founded an AI infrastructure company and sold it[1].
Teaching ML systems at Stanford
At Stanford, Huyen taught CS 329S, a course on machine learning systems design[1][2]. The course page lists her as the instructor, with Christopher Ré among its advisors[2]. The course defined machine learning systems design as the process of defining the software architecture, infrastructure, algorithms and data for an ML system so that it meets specified requirements[2]. Students built a machine learning application as a final project, which made up most of the grade, and presented their work at a demo day[2].
The lecture notes for the course were expanded into her book Designing Machine Learning Systems, published by O'Reilly in 2022[2][3]. The book takes a holistic approach to designing ML systems and looks at each design decision, such as how to create training data, which features to use, how often to retrain models and what to monitor, in terms of how it helps the whole system meet its goals[3]. Her website says it has been translated into more than ten languages[1][3].
AI Engineering
Huyen's book AI Engineering was published by O'Reilly in 2025[3]. It covers the process of building applications on top of readily available foundation models, and it sets out a practical framework for developing and deploying an AI application[3]. The book's GitHub repository says it is not a tutorial book and has few code snippets[4]. Instead it addresses questions such as how to evaluate an application, what causes hallucinations, how retrieval-augmented generation works, how to build and evaluate an agent, and when to finetune a model[4].
The repository lists translations into languages including Japanese, Korean, Chinese, Vietnamese, Spanish, French, German and Polish[4]. Huyen writes on her website that AI Engineering was the most read book on the O'Reilly platform in 2025[1]. She also wrote Machine Learning Interviews (2021), a free and open-source book for job candidates[3].
Writing on AI systems and agents
Huyen writes about AI system design on her blog[1]. In July 2024 she published Building A Generative AI Platform, which outlines the common components she saw in companies' generative AI platforms[5]. The post builds up from a simple architecture by adding context construction, guardrails, a model router and gateway, caching, agentic workflows and observability[5].
In January 2025 she published Agents, adapted from the agents section of AI Engineering[6]. The post covers the tools an agent can use, how foundation models plan, and the ways agents fail, along with how to evaluate them[6]. She notes in the post that AI agents are an emerging field without established theoretical frameworks[6].
She has also written four Vietnamese books of stories about people she met around the world[1]. She builds small public tools, including GoodAIList for discovering open-source AI projects and lazynlp for building large text datasets[1].
Known for
- AI Engineering (O'Reilly, 2025)
- Designing Machine Learning Systems (O'Reilly, 2022)
- Teaching CS 329S at Stanford
- Writing on AI agents and generative AI platforms
Sources
- Chip Huyen: personal website — checked 6 Nov 2026
- Stanford CS 329S: Machine Learning Systems Design, course page — checked 6 Nov 2026
- Chip Huyen: Books — checked 6 Nov 2026
- GitHub: AI Engineering book repository — checked 6 Nov 2026
- Chip Huyen: Building A Generative AI Platform, July 2024 — checked 6 Nov 2026
- Chip Huyen: Agents, January 2025 — checked 6 Nov 2026