frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Pledging Another $400k to the Zig Software Foundation

https://mitchellh.com/writing/zig-donation-2026
340•tosh•1h ago•86 comments

Never Give Them Your Face

https://nevergivethemyourface.com/
276•audiodude•1h ago•160 comments

Claude Code's "extended thinking" is a summary- not authentic thinking

https://patrickmccanna.net/the-text-in-claude-codes-extended-thinking-output-is-not-authentic/
71•0o_MrPatrick_o0•1h ago•50 comments

Moebius: 0.2B image inpainting model with 10B-level performance

https://hustvl.github.io/Moebius/
44•DSemba•1h ago•6 comments

Deno Desktop

https://docs.deno.com/runtime/desktop/
776•GeneralMaximus•10h ago•307 comments

Codex logging bug may write TBs to local SSDs

https://github.com/openai/codex/issues/28224
299•vantareed•8h ago•166 comments

GLM 5.2 vs. Opus

https://techstackups.com/comparisons/glm-5.2-vs-opus/
337•ritzaco•8h ago•241 comments

Chevron signs 20-year power agreement with Microsoft for West Texas data center

https://www.chevron.com/newsroom/2026/q2/chevron-signs-20-year-power-agreement-with-microsoft-for...
23•cdrnsf•1h ago•18 comments

Die analysis of the 8087 math coprocessor's fast bit shifter

https://www.righto.com/2020/05/die-analysis-of-8087-math-coprocessors.html
18•Jimmc414•1h ago•5 comments

Show HN: Got sick of ads, so I made my own logic puzzle site

https://puzzlelair.com/
52•HaxleRose•3h ago•35 comments

I built Ponytrail, a local audit trail for AI coding-agent edits

https://github.com/0xroylee/ponytrail
14•1997roylee•45m ago•6 comments

DHL Set to Transport Goods on New Wind-Powered Cargo Ships

https://www.wsj.com/pro/sustainable-business/dhl-set-to-transport-goods-on-new-wind-powered-cargo...
29•julienchastang•44m ago•4 comments

Help I accidentally a wigglegram

https://lmao.center/blog/wiggle-accidents/
395•gregsadetsky•2d ago•93 comments

Did my old job only exist because of fraud?

https://david.newgas.net/did-my-old-job-only-exist-because-of-fraud/
733•advisedwang•17h ago•330 comments

Apertus – Open Foundation Model for Sovereign AI

https://apertvs.ai/
490•T-A•18h ago•165 comments

Munich 1991: The Roots of the Current AI Boom

https://people.idsia.ch/~juergen/ai-boom-roots-munich-1991.html
163•tosh•2d ago•70 comments

Maria Isabel Sánchez Vegara on Her 100th "Little People, Big Dreams" Book

https://www.amightygirl.com/blog?p=36753
20•zeristor•2d ago•1 comments

Why Drawing Tablet Brands Won't Collaborate on Linux Floss Drivers

https://www.davidrevoy.com/article1154/why-drawing-tablet-brands-wont-collaborate-on-linux-floss-...
113•Tomte•3h ago•42 comments

There is minimal downside to switching to open models

https://www.marble.onl/posts/cancel_claude.html
330•amarble•18h ago•277 comments

Investors get real-time view of UK bond market activity for the first time

https://www.fca.org.uk/news/press-releases/investors-get-real-time-view-uk-bond-market-activity-f...
79•monkeydust•8h ago•53 comments

Nvidia Halos

https://www.nvidia.com/en-us/ai-trust-center/halos/autonomous-vehicles/
44•ilreb•1h ago•21 comments

Manticore Search 27.1.5: Auth, sharding, conversational and faster vector search

https://manticoresearch.com/blog/manticore-search-27-1-5/
30•snikolaev•5h ago•0 comments

Alan Greenspan Dies at 100; Led Fed During Boom Before 2008 Bust

https://www.bloomberg.com/news/articles/2026-06-22/alan-greenspan-dies-at-100-led-fed-during-boom...
90•helsinkiandrew•4h ago•62 comments

Granularity comes at a cost – Game Theory

https://www.sidhantbansal.com/2026/Granularity-comes-at-a-cost/
8•sidhantbansal•2d ago•0 comments

Sakana Fugu

https://sakana.ai/fugu/
177•Finbarr•13h ago•102 comments

My 1992 view of the problems of computer programming in 1992

https://blog.plover.com/prog/fortran-i.html
75•speckx•3d ago•34 comments

Rive, Fast and reliable background jobs in Go

https://github.com/riverqueue/river
13•mountainview•5h ago•0 comments

Memory Safe Inline Assembly

https://fil-c.org/inlineasm
150•pizlonator•2d ago•36 comments

Good results fine tuning a local LLM like Qwen 3:0.6B to categorize questions

https://www.teachmecoolstuff.com/viewarticle/fine-tuning-a-local-llm-to-categorize-questions
191•dev-experiments•16h ago•36 comments

Everything is logarithms

https://alexkritchevsky.com/2026/05/25/everything-is-logarithms.html
273•E-Reverance•18h ago•56 comments
Open in hackernews

Show HN: Got sick of ads, so I made my own logic puzzle site

https://puzzlelair.com/
52•HaxleRose•3h ago

Comments

pavel_lishin•1h ago
This is nice!

Readers may also enjoy Simon Tatham's puzzle collection, available for mobile as well: https://www.chiark.greenend.org.uk/~sgtatham/puzzles/

(My favorite currently is Dominosa. Playing the Hard mode is teaching me new patterns.)

penr0se•1h ago
Mine is currently Net, 7x7 grid with wrapping variant. I take about 5 minutes on average to solve a level, which is the sweet spot for me
abhashanand1501•1h ago
How do you plan to keep it free from ads? I see it has accounts, if it was completely client side, it could have been forever free.
HaxleRose•1h ago
It doesn't cost me much to run. About $9/month for the VPS + domain and I run other apps on the same VPS. Just my little hobby server. It does have monetization though, but I give the first 25 puzzles of each type + difficulty for free and then you pay a few bucks to unlock the rest. Maybe I'm being too generous, but it's not a big deal really.
GL26•1h ago
Please copy the LinkedIn Games !
HaxleRose•1h ago
Oh yeah? I'll have to check them out. Good tip!
lewistaariq•49m ago
I second this suggestion.
iooi•1h ago
Nice job! I made something similar, mostly for myself: https://www.vexling.com

Plan to keep it forever free :)

HaxleRose•1h ago
Nice! real cool! This site does have monetization, but I give the first 25 puzzles free for each difficulty level. So a good amount of free content and I'm still trying to add more games.
underyx•1h ago
I feel like the Nonogram is AI generated? There’s no way a human would set a perfectly symmetrical “diamond” as a medium difficulty puzzle. Worse yet, the hard difficulty is just “big diamond”, the same thing on a slightly larger grid.
pred_•1h ago
I mean the front page is full of LLM smells, so presumably the games are made that way too.

And that's fair; this whole thing could be one-shot with any of the leading models.

giancarlostoro•50m ago
I one shot games every now and then, just to see how much it can do. For anyone wanting to experiment, I have come to learn that if you make it make browser games the setup is even easier since it can just inject the JS into the HTML and import from a popular CDN, no node, no compilers needed, just a single HTML page with inline JS.
benrutter•44m ago
> the front page is full of LLM smells

I'm curious, What kind of details are you thinking of? I'm not sure I really have much of a radar for LLM websites in the way I do for LLM pictures or music.

shevy-java•38m ago
I don't know for pictures, but I have gotten pretty good at detecting AI in videos. I am noticing these a lot on youtube. Often you can tell, e. g. movements being weird, animals behaving in ways that are only in a short and nowhere else to be found. And some more indicators e. g. youtube insists on showing sexy girls, but the video is clearly "cut" into another video and the surface layers also don't fully align; or some proportions are odd (I don't mean the "regular" ones but e. g. when the biceps looks like semi-hulk, you know something is AI slop). I try to not watch AI slop but sometimes it happens.
vova_hn2•1h ago
By the way, if you are interested in nonograms specifically, there is a great website nonograms.org that has tens of thousands nonograms (both B&W and color) and no ads.
HaxleRose•42m ago
Oh yeah, I think I've seen it before. I got some inspiration for some of the nonograms that I made from some different sites. Not all of the sites that I've seen have had a great UI, especially on mobile, which is why I wanted to add nonograms to this site specifically.
ChrisRR•57m ago
Well i like the theory but all of your nonograms are symmetrical or diamonds. Not exactly much of a puzzle

And what sort of monster doesn't have nonogram sizes in multiples of 5?

HaxleRose•39m ago
Haha, yeah, the early puzzles are pretty simple, but they do start to get more variety as you progress through them.

I hear ya. Maybe multiples of 5 would've been better. Mainly, I was trying to get a good mobile experience with as big of a board as I could. Perhaps not the best call.

inigyou•47m ago
Are you blocking Tor?
HaxleRose•43m ago
Hmmm, if it is, I wasn't aware of it. Let me look into my rate-limiting settings that might block something like that.
HaxleRose•38m ago
Checked on it and I'm not blocking Tor. Are you getting issues? Maybe it's not handling the boost in traffic from this post
neogodless•46m ago
I'm 99% sure this was an invalid Sudoku puzzle.

Just before putting the 2 in here (above the pencil 6), I put in 6 and it said Mistake, so I erased it and put 2. But... why wouldn't 6 be valid there?

https://imgur.com/a/aOnKbiT

EDIT: As per replies, "X" Sudoku is variant with a different rule. While I saw the diagonals "highlighted" in another color, I didn't know that rule. Perhaps it could be added to the page for those unfamiliar with this non-standard Sudoku variant?

EdSchouten•42m ago
Yeah, that Sudoku puzzle has multiple valid solutions, whereas the page only seems to accept a single one.
rmunn•40m ago
There's an extra constraint on that one: the two main diagonals must both have nine different digits. That's what makes it a single-solution puzzle.
doodpants•41m ago
Because there's already another 6 on the same main diagonal. This is an "X Sudoku" puzzle, which means that each main diagonal must have all 9 different numbers.
neogodless•39m ago
All I've never done the diagonals before... oops.

https://imgur.com/a/Dfxf9CJ (before picking 2 / 6 / 7.)

insane_dreamer•36m ago
Nice! A few word games would be cool: scrabble, boggle, etc.
yunruse•36m ago
Nice! I wanted to share a link to Ripple Effect Hard with my time (23:47), but it seems the URL only captures attempts, so there's no real way to link to the puzzle itself.

Might be useful to

- add a wordle-style 'SHARE' button, and/or

- make the canonical URL that of the puzzle (and only the attempt on completing/abandoning it)

SpyCoder77•30m ago
The homepage is a little overwhelming, other than that cool site
ori_b•12m ago
I took a quick look at the source:

      // Fire view events (e.g. unlock_prompt_viewed) for any monetization prompt
      // present in the freshly loaded page. data-analytics-view-events is a JSON
      // array so one rendered prompt can report several events at once.
What's a monetization prompt?
furyofantares•2m ago
Looks great. FYI, Claude has idunno, maybe 20-30 different strongly themed websites it knows how to make, and this newspaper aesthetic is one of them, and all the sites it does this way look exactly the same.

It's a good aesthetic for your site, and I thought it was a good one for one of my sites. But eventually I redesigned my site significantly when I saw that it's gonna be common among vibed-up website designs and they look exactly the same.

benrutter•29m ago
For images, there are some clear styles AI leans heavily on if not actively steered away[0].

It can definitely be prompted pretty successfully though, a bird spotting app was up her on HN recently with some really nice looking woodblock prints that were AI generated (I always feel disappointed/tricked when art turns out to be made by AI, I'm not sure why, it seems to pull the joy out of it for me)

[0] https://lcamtuf.substack.com/p/the-100000-whys-of-ai

HaxleRose•44m ago
I hear ya. Fair criticism. I'm a professional developer myself, but not great at design. I've tried to come up with a different looking site best I could. I went with a newspaper theme like back in the day when you'd get the puzzles in the paper. And then it was my idea to have a sudoku being solved as a graphic on the front page. I would push back that this could be one-shot by any of the leading models including Fable. Each of the 10 puzzle types has to have its own generator and they're different from each other. They have to handle uniqueness, solvability, and difficulty and none of the leading models have nailed even just a single generator on the first shot. Plus, there's monetization, rate limiting, caching, among other things under the hood that models wouldn't typically touch without specific instruction or would, at best, half-ass it. Maybe you have better luck with them, but for my job, I work on a large legacy app as well as various microservices and the LLMs miss things all the time. I have a system I use that does make them perform better, but you still gotta watch em like a hawk.
HaxleRose•50m ago
The nonograms get more difficult as you do them. I actually made the diamond one myself, haha! Not too challenging really. I'm not good with making pixel art, but I probably made half of them by hand and I used Fable 5 to make the rest. I didn't actually find Opus or GPT-5.5 very good at making them. Or if they had an idea that was good, I had to fix it myself. Fable 5 was much better and 80% of its ideas looked decent.
miguel-muniz•38m ago
I was also very confused. I started a medium puzzle and was immediately thrown off by the borders. Thicker borders are usually every 5 cells, but here it looks like they've been added just to equally divide the puzzle into 3 chunks.

Missing small details like these makes it fall into the uncanny valley. It looks like a typical puzzle on the surface but when you try to solve it all the mistakes stick out.