News for hackers. Stuff that matters.

varunneal

Re:Engineers: Stop trying to win other people's game on Tuesday December 09, 2025 @10:33AM (#46206027)
Highly AI generated article, including the whiteboard image.
Re:NanoChat – The best ChatGPT that $100 can buy on Tuesday October 14, 2025 @12:16PM (#45581838)
I think it's good to bring up Bernstein & Newhouse as well as Yuchen Jin, Jiacheng You and the other speedrunners who helped iterate on Muon. But I think it's very fair to call Keller Jordan the main author of Muon of its current form. I'm also in the speedrunning community though maybe not as long as you have
Re:NanoChat – The best ChatGPT that $100 can buy on Monday October 13, 2025 @04:53PM (#45573236)
Muon was invented by Keller Jordan (and then optimized by others) for the sake of this speedrunning competition. Even though it was invented less than a year ago, it has already been widely adopted as SOTA for model training
Re:Try and on Monday August 11, 2025 @10:26AM (#44864476)
I think this is a different construction. "And" in "try and find out" is acting as a conjunction, separating two different actions
Re:Building MCP servers for ChatGPT and API integrations on Thursday July 24, 2025 @05:44PM (#44676637)
this guide is just for an example of how to build a single mcp (e.g. a vector store). chatgpt connectors implement mcps in general now
claude sonnet typically forgets about uv script syntax in my experience. I usually find myself having to paste in the docs every time. By default it wants to use uv project syntax.
Re:Tetrachromatic Vision on Tuesday June 17, 2025 @05:58PM (#44304411)
Not publicly, but a few people in berkeley are working on it. Here is a paper from last year: https://imjal.github.io/theory-of-tetrachromacy. (Disclaimer: i am on this paper).

They've prototyped displays that can test for it as well.

Re:Gemini-2.5-pro-preview-06-05 on Thursday June 05, 2025 @01:33PM (#44193837)
Have you tried o3 on those problems? I've found o3 to be much more impressive than Opus 4 for all of my use cases.
isn't this survivorship bias? e.g. people who genuinely feel highly nihilistic, that there is no order, structure, meaning, etc. are very unlikely to be successful--and also unlikely to continue choosing to be alive
Re:Alasdair MacIntyre Had Died on Friday May 23, 2025 @10:37AM (#44073306)
How can reason give us shared values across cultures? Why do you suppose your moral clarity was generated from your brain and not your gut?
Re:Claude 4 on Thursday May 22, 2025 @04:17PM (#44066404)
Why is o3-mini there but not o3?
Re:Teachable Machine on Tuesday May 20, 2025 @01:16PM (#44043779)
I believe this is from 2019 or 2020. Any LLM e.g. Gemini could one shot any problem here, I believe.
Re:PDF to Text, a Challenging Problem on Wednesday May 14, 2025 @10:29AM (#43984987)
I'll check it out!
Re:Data Centers in Space on Tuesday May 13, 2025 @06:13PM (#43978343)
I'll take the long side on that bet
Re:PDF to Text, a Challenging Problem on Tuesday May 13, 2025 @02:40PM (#43976203)
I've been hacking away at trying to process PDFs into Markdown, having encountered similar obstacles to OP regarding header detection (and many other issues). OCR is fantastic these days but maintaining a global structure to the document is much trickier. Consistent HTML seems still out of reach for large documents. I'm having half-decent results with Markdown using multiple passes of an LLM to extract document structure and feeding it in contextually for page-by-pass extraction.
hey there's preview
Re:Python's new t-strings on Monday April 21, 2025 @01:37PM (#43754430)
Would be interested in inclusion of PEP 292 [1] in your discussion here, which introduced `string.Template`. Is this Template going to be deprecated?

[1] https://peps.python.org/pep-0292/

Re:OpenAI Audio Models on Thursday March 20, 2025 @01:38PM (#43426306)
One of the most novel demos I've seen openai ship in a few years. I love how it looks almost like a synth. Fun to play around with!
Cool anecdote from inside article

> A seasoned performance engineer was looking through Strobelight data and discovered that by filtering on a particular std::vector function call (using the symbolized file and line number) he could identify computationally expensive array copies that happen unintentionally with the ‘auto’ keyword in C++.

> The engineer turned a few knobs, adjusted his Scuba query, and happened to notice one of these copies in a particularly hot call path in one of Meta’s largest ads services. He then cracked open his code editor to investigate whether this particular vector copy was intentional… it wasn’t.

> It was a simple mistake that any engineer working in C++ has made a hundred times.

> So, the engineer typed an “&” after the auto keyword to indicate we want a reference instead of a copy. It was a one-character commit, which, after it was shipped to production, equated to an estimated 15,000 servers in capacity savings per year!

to support custom key mappings, as many people with fancy ergonomic keyboards use
very beautiful simple animation. Good visualization of the proof of the algo; namely, that at each step i of the outer loop the first i elements are sorted (this is easy to show via induction)
Re:The Sudoku Affair on Wednesday February 05, 2025 @04:41PM (#42955595)
Hilarious post. I recommend people read to the end

> If Jeffries started with a different core representation, then it's likely his subsequent design decisions would also change. The bookkeeping for constraint propagation might push him towards Norvig's relational approach to the rules of Sudoku; rather than continually recomputing the rows, columns, and boxes, he could simply have a map of each cell onto its peers. He could distill every lesson of the previous posts, creating something simpler and faster.

> But Jeffries isn't in the business of starting over. He not only believes in incremental design, but in using the smallest possible increments. In his posts, he regularly returns to GeePaw Hill's maxim of "many more much smaller steps." He is only interested in designs that are reachable through a series of small, discrete steps.

Jeffries has radicalized me. This sort of puttering-around with "incremental design" is too pervasive in the corporate world. In software we have the luxury of rethinking from first principles, and we must use it. Death to MMMSS

Re:Analysis of 76,882 Product Hunt Products, from 2014 – 2021 on Thursday January 30, 2025 @11:33AM (#42879384)
I really like the thesis of this post: that there is a fundamental divide in tech that can be summarised as "the gamer" and the productivity-maxxing "nihilist". Here's how they describe gamers:

> In this sense, the gamer does not only include those who play video games (nor does it necessarily include all those who do). The gamer is a synecdoche; skateboarders, people who seriously play sports, interested music listeners, cinephiles, chess players, big wave surfers, and anyone else who orients their lives towards experiences that contain fun as a major ingredient are all subsumed into the persona. Casually playing a game on a subway or after a busy day to unwind doesn't by itself garner the gamer moniker, while taking paid vacation time to spend a focused 100 hours on Elden Ring does.

I think at the heart of tech, the pure hackers, and basically all great things come from these people. They are the people who take what they do seriously, and strive for perfection even where it is not utilitarian. It's how you avoid slop, and build great things...

To some extent I think the labels "gamer" and "nihilist" are overly generous. The divide is "people who love the tech" and "grifters". Creating video game is the least financially profitable venture for an expert software engineer, but (because of this) it is the most spiritually fulfilling. Every gamer knows the indie project made with love beats the billion dollar AAA project 9/10 times.

norepinephrine = primary neurotransmitter for the sympathetic nervous system (fight or flight). Mostly produced in brain stem/adrenal glands.
Your prompt verbatim turned out quite well, single-shot.

https://claude.site/artifacts/0f696bf8-399d-42c3-93c0-296493...

so beautiful. I can imagine using this a lot if some buttons are added (start/stop harmony, view note/resonance, etc). great job