Top all-time · #635
Chip Huyen
Personal blog · huyenchip.comaiprogrammingcareer
Chip Huyen's personal blog at huyenchip.com covers ai, programming and career. Since 2017, 98 different Hacker News readers have linked it 124 times, most often “RLHF: Reinforcement Learning from Human Feedback”. It is still writing: the last post was on 2025-01-16, and you can follow it by feed.
Sometimes recommended · 14 of 36 sampled comments recommend it?
Why HN links it
HN readers link Chip Huyen's blog for its posts on RLHF, LLM engineering and MLOps, and her ML interview book, which commenters recommend for AI engineering.
- Recommended for · 14 comments
- Called the RLHF post 'exhaustive' and 'great'; LLM engineering post 'highly recommended'; interview book 'practical'.
Summarised by Claude from a sample of 36 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
I work to bring AI into production. I write about AI system design.
Most-linked posts?
- RLHF: Reinforcement Learning from Human Feedback12 people · 16 mentions · first linked 2023-05-03
- Real-time machine learning: challenges and solutions8 people · 8 mentions · first linked 2022-01-02
- Agents – Adapted from the AI Engineering Chapter by Chip Huyen5 people · 6 mentions · first linked 2025-01-07
- Multimodality and Large Multimodal Models (LMMs)5 people · 5 mentions · first linked 2023-10-11
- Building LLM Applications for Production5 people · 5 mentions · first linked 2023-04-12
- Machine learning is going real-time5 people · 5 mentions · first linked 2020-12-28
- Data Distribution Shifts and Monitoring4 people · 5 mentions · first linked 2022-02-08
- A friendly introduction to machine learning compilers and optimizers4 people · 4 mentions · first linked 2021-09-08
- Building a Generative AI Platform4 people · 4 mentions · first linked 2024-07-25
- Introduction to Machine Learning Interviews Book4 people · 4 mentions · first linked 2021-06-22
What Hacker News says?
> Yes. It's not a written definition. It's a living and changing definition created by the people who work there and how YOU make them feel. You need to use your people skills in order to extract the definition from your co-workers. Agreed, but that's my very complaint! As they say, the pendulum has swung too far! (In my opinion, of course.) > Look Mom! We are freely debating something right here, right in society!…
Comment linking /2017/08/09/sexism-in-silicon-valley.html
on “Outraged about the Google diversity memo?”Submitted What I learned from looking at every AI/ML tool I could find as a story · 459 points
HN discussionThis is good also https://huyenchip.com/2023/05/02/rlhf.html
Comment linking /2023/05/02/rlhf.html
on “RLHF Book”Submitted What I learned from looking at 900 most popular open source AI tools as a story · 413 points
HN discussionHi, I'm Chip, the author of the post. I spent waaaay too much time doing research for this. The AI engineering layer was especially hard because so many tools have similar and/or overlapping features. It was also a lot of pain trying to understand the repos that only have Chinese in the README do. The full list of the repos is published here: https://huyenchip.com/llama-police
Comment linking /llama-police
on “What I learned from looking at 900 most popular open source AI tools”Submitted How I, a woman in tech, benefited from sexism in Silicon Valley as a story · 321 points
HN discussionThis section invites some discussion (says British bloke): " https://huyenchip.com/2024/03/14/ai-oss.html#the_growing_chi... "
Comment linking /2024/03/14/ai-oss.html
on “What I learned from looking at 900 most popular open source AI tools”Submitted Analysis of compensation, level, and experience details of 19k tech workers as a story · 275 points
HN discussionSome youtube channels are good too. https://www.youtube.com/@umarjamilai https://huyenchip.com/blog/
Comment linking /blog/
on “Ask HN: Recommendation for a SWE looking to get up to speed with latest on AI”Submitted Confession of a so-called AI expert as a story · 203 points
HN discussionYou can give this article by Chip Huyen a read. Mayhaps you will find a niche for a solo or small dev team. Though it is focused on MLOps if that makes a different for the type of niche you're looking for. https://huyenchip.com/2020/06/22/mlops.html
Comment linking /2020/06/22/mlops.html
on “The cost to train an AI system is improving at 50x the pace of Moore’s Law”Submitted Data scientists shouldn’t need to know Kubernetes as a story · 181 points
HN discussionIs she having doubts brought on by the recent Google Manifesto perhaps: http://huyenchip.com/2017/08/09/sexism-in-silicon-valley.htm... posted earlier.
Comment linking /2017/08/09/sexism-in-silicon-valley.html
on “Confession of a so-called AI expert”Submitted Introduction to Machine Learning Interviews Book as a story · 151 points
HN discussions/thinking about what he's describing/thinking about what she's describing https://huyenchip.com/
Comment linking /
on “Confession of a so-called AI expert”Submitted Measuring personal growth as a story · 109 points
HN discussionFor SWEs interested in "AI Engineering" (either getting involved in how models work, or building applications on them), there's a critical paradigm shift in that using "AI" requires more of an experimental mindset than software engineering typically does. - I strongly recommend Chip Huyen's books ("Designing Machine Learning Systems" and "AI Engineering") and blog ( https://huyenchip.com/blog/ ). - Andreessen Horowi…
Comment linking /blog/
on “Ask HN: Go deep into AI/LLMs or just use them as tools?”Submitted A friendly introduction to machine learning compilers and optimizers as a story · 72 points
HN discussionYes. The learning comes from running tests on the program and ensuring they pass. So running as an agent. Tests and compiler give hard feedback- thats the data outside the model that it learns from. I think modern RLHF schemes have models that train LLMs. LLMs teaching each other isn't new. My knowledge is limited, just based on a read of https://huyenchip.com/2023/05/02/rlhf.html though.
Comment linking /2023/05/02/rlhf.html
on “Designing a Language (2017)”Submitted Sampling for Text Generation as a story · 50 points
HN discussionThere is also the much larger confusion arising of whether it's an ML compiler in the sense of being a specialized compiler for ML code (compiles a description of a network to efficient ML-specific low-level IR to enable further optimization and code generation), or whether it's an ML compiler, a compiler (or part of it) that uses ML. The first meaning is extremly common, the top 2 results for "ML Compiler" on googl…
Comment linking /2021/09/07/a-friendly-introduction-to-machine-learning-compilers-and-optimizers.html
on “Google ML Compiler Inlining Achieves 3-7% Reduction in Size”Submitted Introduction to Machine Learning Interviews Book as a story · 27 points
HN discussion"RTFM"[0] as they say [0] https://huyenchip.com/ml-interviews-book/contents/3.1.1-base...
Comment linking /ml-interviews-book/contents/3.1.1-base-salary.html
on “Introduction to Machine Learning Interviews Book”Submitted Real-time machine learning: challenges and solutions as a story · 23 points
HN discussionIt's just a fancy word for clamping the new reward value to within some delta of the original value. Otherwise the model ends up "exploiting" outliers that make sense to machines but not to humans. They do the same thing with PPO in RLHF. Great article, if you're interested: https://huyenchip.com/2023/05/02/rlhf.html#3_2_finetuning_us...
Comment linking /2023/05/02/rlhf.html
on “Offline Reinforcement Learning for LLM Multi-Step Reasoning”Submitted Real-time machine learning: challenges and solutions as a story · 10 points
HN discussionI haven't read these all the way through myself but I've seen enough of them that I'm confident suggesting them: - Prompt Engineering for LLMs by John Berryman and Albert Ziegler: https://www.amazon.com/Prompt-Engineering-LLMs-Model-Based-A... - AI Engineering by Chip Huyen, which I recommend based on the strength of this extract about "agents": https://huyenchip.com/2025/01/07/agents.html
Comment linking /2025/01/07/agents.html
on “AI isn't going to kill the software industry”Submitted Data Distribution Shifts and Monitoring as a story · 7 points
HN discussionI was a MLE Tech Lead at Snap and laid some of the foundations of the generative AI infra at Snap. I would highly recommend MLE route as a very rewarding career path. This book is a very good introduction to designing Machine Learning Systems for production: https://www.amazon.com/Designing-Machine-Learning-Systems-Pr... This blog by the same author is highly recommended as an intro into building production grade AI…
Comment linking /2023/04/11/llm-engineering.html
on “Ask HN: How to pivot to a Machine Learning engineer?”Submitted 7 Reasons not to join a startup and 1 reason to as a story · 7 points
HN discussion
