Top all-time · #999
Eugene Yan
Personal blog · eugeneyan.comaidataprogramming
Eugene Yan's personal blog at eugeneyan.com covers ai, data and programming. Since 2020, 69 different Hacker News readers have linked it 185 times, most often “Text-to-Image: Diffusion, Text Conditioning, Guidance, Latent Space”. It is still writing: the last post was on 2026-06-21, and you can follow it by feed.
I work to bridge the field and the frontier, and help build safe, reliable AI systems that scale. I also write and speak about AI-powered systems and products, and prototype for fun.
Most-linked posts?
- Text-to-Image: Diffusion, Text Conditioning, Guidance, Latent Space7 people · 8 mentions · first linked 2022-11-29
- Task-specific LLM evals that do and don't work6 people · 6 mentions · first linked 2024-04-01
- System Design for Recommendations and Search5 people · 5 mentions · first linked 2021-06-28
- Evaluating the Effectiveness of LLM-Evaluators (a.k.a. LLM-as-Judge)4 people · 6 mentions · first linked 2024-08-21
- Patterns for Building LLM-Based Systems and Products4 people · 5 mentions · first linked 2023-08-02
- A Spark of the Anti-AI Butlerian Jihad (On Bluesky)4 people · 4 mentions · first linked 2024-12-11
- Red Flags to Look Out for When Joining a Data Team4 people · 4 mentions · first linked 2022-02-17
- How to Generate and Use Synthetic Data for Finetuning4 people · 4 mentions · first linked 2024-02-11
- Uncommon Uses of Python in Commonly Used Libraries4 people · 4 mentions · first linked 2022-08-09
- How to Work and Compound with AI4 people · 4 mentions · first linked 2026-05-05
What Hacker News says?
Hi, Hamel here. I'm one of the co-authors. I'm an independent consultant and not all clients allow me to talk about their work. However, I have two that do, which I've discussed in the article. These are two production use cases that I have supported (which again, are explicitly mentioned in the article): 1. https://www.honeycomb.io/blog/introducing-query-assistant 2. https://www.youtube.com/watch?v=B_DMMlDuJB0 Othe…
Comment linking /writing/
on “What we've learned from a year of building with LLMs”Submitted Stop Taking Regular Notes; Use a Zettelkasten Instead as a story · 785 points
HN discussionHi, Hamel here. I'm one of the co-authors. I'm an independent consultant and not all clients allow me to talk about their work. However, I have two that do, which I've discussed in the article. These are two production use cases that I have supported (which again, are explicitly mentioned in the article): 1. https://www.honeycomb.io/blog/introducing-query-assistant 2. https://www.youtube.com/watch?v=B_DMMlDuJB0 Othe…
Comment linking /writing/
on “What we've learned from a year of building with LLMs”Submitted The First Rule of Machine Learning: Start Without Machine Learning as a story · 764 points
HN discussion100% agree with Ted's take. One of the authors wrote about splitting up prompts here too: https://eugeneyan.com/writing/prompting/#split-catch-all-pro...
Comment linking /writing/prompting/
on “What We Learned from a Year of Building with LLMs”Submitted Improving recommendation systems and search in the age of LLMs as a story · 408 points
HN discussionLots of people can get impressive demos up and running, but if you want to run AI products in production, you're going to have to do system evals. System evals make sure your product is doing what it says on the box with unquantifiable qualities. We wrote a zine on system evals without jargon: https://forestfriends.tech Eugene Yan has written extensively on it https://eugeneyan.com/writing/evals/ Hamel has as well.…
Comment linking /writing/evals/
on “Ask HN: Recommendation for a SWE looking to get up to speed with latest on AI”Submitted What we've learned from a year of building with LLMs as a story · 401 points
HN discussionRead through this making flashcards as you to: https://eugeneyan.com/writing/llm-patterns/ Then spin up a RAG-enhanced chatbot using pgvector on your favourite subject, and keep improving it when you learn about cool techniques
Comment linking /writing/llm-patterns/
on “Ask HN: Recommendation for a SWE looking to get up to speed with latest on AI”Submitted Simplicity is an advantage but sadly complexity sells better (2022) as a story · 376 points
HN discussionDoes this handle real time recommendations as described here: https://eugeneyan.com/writing/real-time-recommendations/
Comment linking /writing/real-time-recommendations/
on “Show HN: Gorse – An Out-of-the-box open-source Recommender System”Submitted Writing better by answering why, what, how as a story · 278 points
HN discussionI don't know. On one hand, it is true we are the last line of defense. I do wish more of us said no to putting tracking on websites, among other things. But on the other hand, a random software engineer's sense of ethics might be different than you. Do we want activist software engineers? I mean, field of study changes perspective. Recently, there was that backlash against a release of a 1-mil post dataset of bluesk…
Comment linking /writing/anti/
on “Don't Get Distracted (2017)”Submitted Most of the time, you don’t really need another MOOC as a story · 194 points
HN discussionI think that will be very possible soon! We continue to write about it publicly :) Also thanks to my friends and colleagues who write a lot on this subject that I frequently collaborate with: - Shreya Shankar https://www.sh-reya.com/ - Eugene Yan https://eugeneyan.com/ - Bryan Bischof https://bio.site/Docdonut
Comment linking /
on “About AI Evals”Submitted Task-specific LLM evals that do and don't work as a story · 182 points
HN discussionThere are actually some clever approaches to eval abstractive summarization. Examples: https://eugeneyan.com/writing/evals/#summarization-consisten...
Comment linking /writing/evals/
on “Mayo Clinic's secret weapon against AI hallucinations: Reverse RAG in action”Submitted Uncommon Uses of Python in Commonly Used Libraries as a story · 178 points
HN discussionI second the request. I think communication is a drastically underappreciated skill even though we have all suffered due to poor communication professionally (undocumented libraries, poorly specified requirements, an inability to generate buy-in for our ideas). I'll throw in a few recommendations I have from some recent research at work: https://eugeneyan.com/writing/writing-docs-why-what-how/ https://eugeneyan.com/…
Comment linking /writing/what-i-do-before-a-data-science-project-to-ensure-success
on “Ask HN: Any great books about technical writing?”Submitted Building the same app using various web frameworks as a story · 161 points
HN discussionhttps://eugeneyan.com/writing/obsidian-copilot/
Comment linking /writing/obsidian-copilot/
on “Ask HN: Has Anyone Trained a personal LLM using their personal notes?”Submitted How I Did It: From Psych Degree to VP of Data Science at Top Startup as a story · 153 points
HN discussionPutting random links in your comments is of limited value, and it shouldn't add authority to your claims. It's actually too common on HN for people to make things up and then put some sort of citation that actually doesn't support what they are claiming whatsoever, knowing that 99.9% of users just don't care that much. The fact that it actually works on some people is hugely unfortunate. However while I'm not the ot…
Comment linking /writing/anti/
on “Bluesky's science takeover: 70% of Nature poll respondents use platform”Submitted Advice for new principal tech ICs (i.e., notes to myself) as a story · 137 points
HN discussionI also find Eugene Yan's less technical blog to be similarly fantastic. https://eugeneyan.com/
Comment linking /
on “What Are Diffusion Models?”Submitted What I wish I knew about onboarding as a story · 115 points
HN discussionReminds me of this: https://news.ycombinator.com/item?id=40266464 https://eugeneyan.com/writing/simplicity/ “Simplicity is a great virtue but it requires hard work to achieve it and education to appreciate it. And to make matters worse: complexity sells better.” — Edsger Dijkstra Especially: Complexity signals more features.
Comment linking /writing/simplicity/
on “The rise of the disposable car”Submitted Push notifications – What to push, what not to push, and how often as a story · 107 points
HN discussionbluesky uniquely hates AI. very toxic around it. https://eugeneyan.com/writing/anti/ its sad that the informal social media bubbles we all have formally forked[0] into Red Twitter and Blue Twitter. just a direct consequence of filters, feeds, algorithms that we say we dont want but vote for with our actions. with Latent Space I try very hard to stay practical and grounded. but this means i lose out to other podcaste…
Comment linking /writing/anti/
on “I am disappointed in the AI discourse”Submitted Uncommon Uses of Python in Commonly Used Libraries (2022) as a story · 99 points
HN discussion
