Chip Huyen is a writer and computer scientist focused on bringing AI systems into production. She wrote Designing Machine Learning Systems and AI Engineering after working on machine-learning tooling at NVIDIA, Snorkel AI and Netflix.[1][2]
Start with Designing Machine Learning Systems companion repository →
Education
Links
Website Books
GitHub 
X LinkedIn Newsletter Google Scholar Background
Huyen’s personal biography says she graduated from Stanford and taught its ML Systems course. Her 2017 teaching correspondence identifies her as a master’s student, which establishes graduate study without supplying a completion date or named degree. She also created and taught Stanford’s TensorFlow for Deep Learning Research course. That teaching background connects to books organized around the decisions engineers face when moving models into applications.[1][3][4]
Career
Her earlier work includes NVIDIA’s NeMo toolkit, where she identifies herself as a core developer, as well as Snorkel AI and Netflix. She also says she founded and sold an AI-infrastructure company. Her present public work centers on writing, AI-system design and small tools. Her portfolio includes lazynlp for building text datasets and small applications exploring open-source discovery and product defensibility.[1]
Contributions
The whole system around a model
Designing Machine Learning Systems treats data processing, feature choices, retraining and monitoring as linked decisions within a production system. Its organizing question is whether those decisions help the overall application meet its objectives. AI Engineering shifts the focus to building with readily available foundation models: choosing models and datasets, evaluating outputs, recognizing application patterns and deploying efficiently. The two books address related stages of practice, from maintaining a learned system to adapting existing model capabilities for a specific product.[2]
Open educational material
Machine Learning Interviews is a free, open-source guide aimed at candidates and also used by hiring managers according to its author. It complements the production books by making interview expectations and preparation explicit. Her books page links repository material for Designing Machine Learning Systems, including chapter summaries and tooling notes, allowing readers to explore the structure before committing to the full text.[2][3]
Tools, writing & talks
Writing, research & talks
Start here
Table of contents, chapter summaries and notes for an iterative approach to production machine learning.
Browse the Designing Machine Learning Systems chapter summaries to identify which production decisions you need to understand. Some familiarity with training a model helps; the guide broadens the question to data and operations.
Chip Huyen covers AI engineering fundamentals, why many companies struggle to ship AI products, how hard coding-agent productivity is to measure, and how to improve apps via user feedback.
Catalog workOriginal sourceThe mistakes that recur when teams build with foundation models, starting with the most common one: reaching for generative AI where it was never the right tool.
Catalog workOriginal source- Agentshuyenchip.com
A treatment of agents in the classical sense, rational agents acting in an environment, and what foundation models change about which agentic applications are finally buildable.
Catalog workOriginal source - Building A Generative AI Platformhuyenchip.com
The components that recur across generative AI platforms once you look at how companies actually deploy them, built up from the simplest possible architecture rather than presented as a finished diagram.
Catalog workOriginal source What a survey of the 900 most popular open-source AI repositories shows about how the ecosystem has changed, four years after the same exercise on the open ML landscape.
Catalog workOriginal sourceInstead of ranking models and picking one, predict which model a user would prefer for a specific prompt, turning human preference data from a leaderboard into a routing signal.
Catalog workOriginal sourceHow generation is actually configured: temperature, top-k and top-p as ways of shaping a probability distribution rather than magic dials, and how test-time compute fits alongside them.
Catalog workOriginal source- Multimodality and Large Multimodal Models (LMMs)huyenchip.com
A primer on multimodality and large multimodal models: why systems that only read text are a poor fit for a world humans navigate by reading, listening and seeing at once, and how such models are built.
Catalog workOriginal source - Open challenges in LLM researchhuyenchip.com
Ten major open research directions in language models, drawn from conversations across industry and academia, with hallucination and context learning the most discussed and multimodality and new architectures the most promising.
Catalog workOriginal source - AI Engineeringhuyenchip.com
A guide to building and deploying applications with existing foundation models, including evaluation and system choices.
Author[2] - Machine Learning Interviewshuyenchip.com
An open guide to machine-learning interview preparation.