frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

France powers down several nuclear reactors due to extreme heat

https://www.lemonde.fr/en/france/article/2026/07/12/france-powers-down-several-nuclear-reactors-d...
2•_Microft•1m ago•0 comments

Grok Build uploading full repos and .envs to GCP

https://twitter.com/xbtoshi/status/2076338252051841512
1•ashleypeacock•2m ago•0 comments

Trump notifies Congress of new war against Iran

https://www.politico.com/news/2026/07/13/trump-notifies-congress-of-new-war-against-iran-00995170
4•bushwart•2m ago•0 comments

AI Agents for Increasing Revenue

https://alum.so/
2•Sumiran7•2m ago•1 comments

The Tick That Hunts Down Its Hosts–Including Us

https://www.newyorker.com/magazine/2026/07/06/the-tick-that-hunts-down-its-hosts-including-us
1•randycupertino•4m ago•0 comments

Venice's access fee doesn't reduce tourism: it selects who can afford it

https://andreafontana.it/en/venice-entry-ticket-overtourism.html
1•trikko•5m ago•0 comments

Microsoft commits $2.5B, 6k employees AI implementation unit

https://www.cnbc.com/2026/07/02/microsoft-commits-2point5-billion-6000-employees-ai-implementatio...
1•rbanffy•5m ago•0 comments

Quantum-Augmented Databases Study Aims to Break Bottlenecks Slowing Data Systems

https://viterbischool.usc.edu/news/2026/07/toward-quantum-augmented-databases-new-usc-study-could...
1•rbanffy•5m ago•0 comments

Brazilian woman held as a slave for 38 years (2021)

https://www.dw.com/en/brazilian-woman-held-as-a-slave-for-38-years/a-56177797
1•toilet•6m ago•0 comments

Human settlement of East Polynesia coincident with prolonged S. Pacific drought

https://www.pnas.org/doi/10.1073/pnas.1920975117
1•janalsncm•8m ago•0 comments

Welcome to the Modern World. Is The Most Unhelpful Thing You Can Say to a Dev

https://codemonkey.studio/welcome-to-the-modern-world-why-that-might-be-the-most-unhelpful-thing-...
1•rmason•9m ago•0 comments

Brain-inspired hardware brings faster, low-power anomaly detection to AI systems

https://techxplore.com/news/2026-07-brain-hardware-faster-power-anomaly.html
1•rbanffy•9m ago•0 comments

I built a domain name discovery tool I don't hate

https://domaame.com
1•cybrjoe•9m ago•1 comments

Meta's AI Glasses Will Activate the Camera Without the Camera Indicator Light

https://www.privacyguides.org/news/2026/07/13/the-next-version-of-metas-ai-glasses-will-activate-...
1•Cider9986•10m ago•0 comments

Evelyn Berezin: instamatic for UAL, then first word processor

https://www.invent.org/inductees/evelyn-berezin
1•stmw•11m ago•0 comments

Apple sues OpenAI after ex-engineer allegedly used bug to steal trade secrets

https://arstechnica.com/tech-policy/2026/07/apple-sues-openai-after-ex-engineer-allegedly-used-bu...
2•bookmtn•12m ago•0 comments

New York City's chief technologist launches team to transform city's technology

https://www.fastcompany.com/91571965/new-york-citys-chief-technologist-is-launching-a-new-team-to...
1•hn_acker•13m ago•1 comments

Department of War Suspends CMMC Phase II Requirements

https://www.war.gov/News/Releases/Release/Article/4542329/forging-the-arsenal-of-freedom-departme...
2•ckrailo•16m ago•0 comments

We gave open models root on broken servers and graded them on the machine

https://glassmkr.com/blog/open-model-ladder-blind-remediation
1•glassmkr•21m ago•0 comments

Making Fable Cheaper Than Opus

https://cognition.com/blog/making-fable-cheaper-than-opus
1•aarvin_roshin•24m ago•0 comments

Harness Engineering for Self-Improvement – Lil'Log

https://lilianweng.github.io/posts/2026-07-04-harness/
1•haritha1313•26m ago•0 comments

Measuring Agents in Production – ICML

https://arxiv.org/abs/2512.04123
1•haritha1313•27m ago•0 comments

Web Client SDK

1•atoapaymentsapp•28m ago•0 comments

Coluber: Language Written in Nim. Speaks Nim, C, Vlang, Python, and JavaScript

https://codeberg.org/distantfar/coluber
1•baranul•30m ago•0 comments

Consumer Resources – CFPB

https://www.consumerfinance.gov/consumer-tools/
1•enjoyyourlife•31m ago•0 comments

V. vulnificus infection has extreme mortality and lacks a treatment protocol

https://en.wikipedia.org/wiki/Vibrio_vulnificus
1•burnt-resistor•31m ago•0 comments

Just Let Me Write Digits

https://gendx.dev/blog/2026/07/13/input-digits.html
1•gendx•31m ago•0 comments

AI buildout poses latest inflation threat

https://apnews.com/article/ai-inflation-federal-reserve-434f02e62a02f9b92e57995d9375df57
3•billybuckwheat•33m ago•1 comments

Show HN: GrindProof – accountability app that roasts your plan-vs-reality gap

https://www.grindproof.co/
1•Codefred•34m ago•0 comments

Apple Plans AirPods-Like Pairing for Meta's Glasses and Quest

https://www.macrumors.com/2026/07/13/apple-plans-airpods-like-pairing-for-meta/
1•nate•34m ago•0 comments
Open in hackernews

Show HN: Super Dario

https://superdario.pawb.de
354•thepasch•3h ago

Comments

starik36•3h ago
Nice and clever!
thatxliner•3h ago
a perfect representation of the situation
m_w_•3h ago
+1 for the claude icon fire bars
GaggiX•3h ago
Someone didn't know how to use the remaining credits.
thepasch•3h ago
The heavy lifting was done by Opus. I wouldn't burn precious soon-to-be-departing Fable usage on this!
andrew_ocs•2h ago
I wouldnt be so sure Fable wouldnt just refuse to do this
thepasch•2h ago
I mean, technically, "I selected Fable for this task" and "Opus did the heavy lifting" aren't mutually exclusive statements...?
forestingfisher•3h ago
Can’t believe this doesn’t already exist.
dhg72•3h ago
There are plenty of critiques of the AI labs but it's a bit ironic leveling criticism in a game which was obviously one-shotted by their models and which would never have been made without AI
androiddrew•3h ago
A model harness can launch a bagillion sub agents and we still call it one shotting.
zahlman•44m ago
... And why shouldn't we? The system still responded to a single prompt, and we still took its first submitted candidate.
ipsum2•3h ago
That's part of the joke.
gordon_freeman•3h ago
who needs this AI-genrated game for real though? Would not I rather play the real Mario instead? this is clearly part of the joke!
croes•2h ago
Fighting them with their own weapons isn’t ironic it’s poetic justice
pixelpanic•3h ago
this guy fits better, I guess: https://helios-i.mashable.com/imagery/videos/04iSw6696vB87oa...
einpoklum•3h ago
This reminded me of how much I suck at side-scroller arcade games. I keep missing those coins and falling off ledges... :-(

Good thing I have a day job. Where I don't use AI.

jake_and_fatman•3h ago
Absolute fire. Love how Deepseek tortoise shell can only be slowed, not killed.
boutell•1h ago
Thank you, I missed that.
Pxtl•50m ago
You can bounce on them forever for infinite coins. But the coins don't matter. Nothing matters.
seydor•2h ago
The game must be banned for 6 months
andrew_ocs•2h ago
I think you should try submitting it to anthropic community :)

https://claude.com/community#:~:text=What,Claude%3F

thepasch•2h ago
I'll... uh, do it next week!
helloplanets•2h ago
Shouldn't the valuation be in Bs instead of Ms?
jagged-chisel•2h ago
Numbers look bigger when multiplied by one thousand
thepasch•2h ago
Grab that coin and you'll get there!
dude250711•2h ago
Like political satire, only in IT.
goldenarm•2h ago
Funny idea but wow the gameplay is terrible and full of bugs. Bad hitboxes, shells that bounce in front of holes, bad physics.

AI still cannot oneshot 2D platformers, the most documented videogame genre with so much source code available.

slopinthebag•2h ago
Not to mention it feels like 30hz on my 120hz monitor. It's actually impressive how it managed to mess that up.

But yeah guys, AI sure will be writing all the code just give it another 6 months...

tavavex•2h ago
It feels like each new SOTA model brings the average quality a bit closer to the asymptote of well-made non-vibecoded software but it can't quite cross it.
nsxwolf•2h ago
Just amazing how right a few people at Nintendo got this in 1985, and its still so easy to get it absolutely wrong.
reactordev•2h ago
Nintendo wasn’t even trying, it was all their old sprite characters from Japan market that didn’t do well smashed together in a rushed title for an NA launch. It surprised them.
Wowfunhappy•1h ago
Huh? This is very much not my understanding of SMB1’s development. Are you maybe thinking of Donkey Kong (made to replace a failed arcade game) and/or the US version of SMB2 (a re-skinned non-Mario Japanese game)?
mrkeen•2h ago
This game was so good, it's as if the developer was given a team of PhD-level experts to work on it.
Waterluvian•1h ago
Haters will suggest AI built this, but I can personally tell there’s at least a kilomolyneux (approx. 55 centikojimas) of PhD-level game dev expertise involved.
thepasch•2h ago
Okay, so, maker here!

I was expecting this to drown in the flood of /newest, so seeing so many folks get a chuckle out of this is a very pleasant surprise. The obvious spark for this was yesterday's second last-minute extension of the Fable promotional window (a phrase comprised of two heavily strained words at this point), and an hour or so of back-and-forth with GLM-5.2 and Opus later, here we are.

The game is, of course, unwinnable on purpose. There is no ending. The flag always escapes, the date always extends, and new GPT releases will continue to wipe your valuation no matter how many coins or funding round power-ups you collect. The win condition is closing the tab because you're tired of it, which I'd argue makes it the most realistic AI-industry simulation currently available on the internet.

The code is, of course, vibe-slopped, most of the imagery is original human-slop (or sourced from icon libraries). Happy to answer any and all questions and take any and all flak. The bugs are a part of the joke, I swear!

humam_alhusaini•2h ago
This is great lol

Can you release a version playable on mobile?

thepasch•2h ago
Yeah, on it right now! Should be up within the hour.

edit: It's up!

kylehotchkiss•1h ago
Level 10 government subsidy invincible mode
thepasch•1h ago
gotta save some ideas for the sequel
Cider9986•2h ago
No mobile support.
parl_match•2h ago
Thanks for your contribution.
thepasch•1h ago
On-screen controls are live!
sherlock-holmes•2h ago
incredible
salahadawi•2h ago

  New Gemini model! Nobody cares.
With the Gemini power-up falling off the screen got a chuckle out of me
mchusma•1h ago
This was my favorite as well.
kaeruct•2h ago
Why do the coins spin faster when I move?
kridsdale1•1h ago
Momentum. And / or Special Relativity.
sajithdilshan•2h ago
The game is really cool, but I'm a bit dissapoint by the music. I want the original them song. Also the fact that the game is never ending bothers my OCD
dancemethis•1h ago
I mean, Sam should be more of a ghoul with those creepy sunken eyes. And Dario that guy with hair on the sides.
meken•1h ago
The music reminds me of Mappy Land (one of my favorite childhood games).
thepasch•1h ago
No credit to me for that! Got it off of here: https://opengameart.org/content/bonus-round-8bit
_s_a_m_•1h ago
hahaahhahahahahahahahahahah
slicendice•1h ago
Absolutely hilarious
CrzyLngPwd•1h ago
I wonder which model built it :-)
neural_thing•1h ago
If you like tech-themed games, I made Good Til Canceled for GTC - check it out - https://goodtilcanceled.com/
ryandrake•1h ago
Desktop Safari 18.6 / Mac: Nice, but none of the text on the signs are readable. Just appears as unreadable dots. Also the music doesn't work on Safari.

Firefox works great!

thepasch•1h ago
Agh, sorry about that. I had no Mac around to test this on, though I might be able to get one by proxy. I'll try to get it fixed.
kridsdale1•1h ago
Same bug in iOS 27 chrome.
aliasxneo•1h ago
Wow, my entire feelings on this series of events summed up in a single game. Extra rich its vibe coded. We need more of this.
clutter55561•1h ago
This wins the internet this week!
jdw64•1h ago
After playing this game, I think it's time to launch 'vibe game design.' Now that the code is solved, it's time to release 'vibe game design,' 'vibe music design,' and 'vibe control design' for serious game design
danielrmay•55m ago
For all of the advertising of GPT 5.6 in this charming little game, the chatgpt.com UI has failed to load in the past several hours for me
geoffbp•53m ago
Open source it? :-)
nico•47m ago
Omg, so funny! I can’t believe I played this for like 5 min straight

Are those little Sam Altmans?!

Love the power ups: tokenmaxxing, usage reset

reactordev•1h ago
Nope, SMB1 was a smash up. SMB2 was made by folks that had no idea what they were doing so SMB3 went back to basics. I very much remember the sequence of events. Mario Bro's (and Donkey Kong) were games on the Famicom in Japan that were decent. The issue was with Arcade's in Japan making more money than home systems so they introduced the Nintendo VS cabinets. This helped but it wasn't enough and so in 1985 they greenlit the Super Mario Bros proposal while dealing with the NA launch and the NYC event. Shigeru Miyamoto carried the development and he deserves the credit but corporate was throwing shit at the fan and seeing what sticked. He just so happened to have a plan that required less work from creative.
uncivilized•42m ago
Any links to read more about this?
nsxwolf•1h ago
Shigeru Miyamoto has spoken many times about the process of perfecting the controls for SMB, which is quite the opposite of "wasn't even trying".
reactordev•1h ago
Oh he absolutely cared about the programming and process of game development, don't get me wrong. It's that Nintendo (corporate) needed something for the NA market and he was eager to fill that gap with already used characters (such as Mario). Anything else is corporate propaganda they have been telling as a creation myth to make it sound more fairytale-like than it was. They were going broke. They needed money. North America was ripe for the picking. The risk paid off.
input_sh•2h ago
On the "bright side", it's impossible to die!
disillusioned•2h ago
That's the power of venture capital!
miyoji•2h ago
This was my favorite bit of meta-commentary: it doesn't matter how poorly you play, you never lose!
kridsdale1•1h ago
Even survived the Hegseth Bullet Bill attack
thisisit•1h ago
More like keeps "extending by 1 more week"
mkagenius•1h ago
you mean amodei
zahlman•48m ago
It seems that you do lose a life when hit by an enemy or falling in a pit, but the latter freezes until you click the toggle sound button, and running out of lives resets the counter to 3 and halves your "valuation".
reactordev•2h ago
You’re absolutely right! /s
vunderba•2h ago
Yep. Additionally having experimented pretty heavily in this space, I'll state that even with state-of-the-art models (like NB Pro or GPT-Image-2), good luck getting them to generate sprite sheets with consistent frames for animation particularly if you're shooting for a pixel art aesthetic.

I've made a few sprite-based animations with AI assistance but it still required a fair amount of manual photoshop (or in my case Aseprite) post-processing.

nananana9•1h ago
If you're not going for pixel art, use the video to video models (the character swap ones that take in a video reference of the animation and a still of the character (the tiktok deepfake models)).

You can e.g. download a run cycle from Mixamo, use Blender to render a video from whatever angle you wish, and feed that into the model along with your sprite on a green background, it's pretty much perfect.

throwthrowuknow•56m ago
It’s a tough problem but Retro Diffusion is making some incredible progress

https://retrodiffusion.ai/

dvt•2h ago
One-shotting is a bit of a red herring, imo. Let's say I wanted to build a "Super Dario" meme platformer. I would do some research, find a platformer on Codepen or Github that "feels" good, tweak it to my liking, change the sprites, and voila.

I constantly have to tell agents to just "look it up online" and "don't hallucinate your own components" because people have already done this a million times. Ironically, being more lazy could make these models more useful.

bogwog•1h ago
So you're saying that it's not fair to rate a coding model on its ability to code, and instead the best way to use it is to tell it to find existing human-written code online rather than generate actual code on its own?
darkwater•1h ago
Isn't part of the human coding that Super Dario tells us LLM are going to replace also made of what you are saying GP is saying?
dvt•31m ago
If you're testing how good LLMs are at compressing information, then I think that's a fair test. Personally, I don't really think that's where their strength comes from (especially considering how much more useful local models that are orders of magnitude smaller than Claude/OpenAI-tier models have gotten). In other words, we already have a "super-intelligence"—it's called the internet, so just use the darn thing.
sefrost•1h ago
Yes for any slightly non-standard web UI components I either tell the LLM to search the web and find examples or I provide them myself.

Usually the ones I provide myself are better than the ones they find themselves.

I'm not super clear how they search the web or what they do and don't have access to. It seems to be quite fast.

jimkleiber•53m ago
TBH, I wonder how much one-shotting posts are just lying about one-shotting it.
uncivilized•45m ago
Many for sure.
micromacrofoot•1h ago
I think that supports the premise more than it defies it
boutell•1h ago
Strong agree. Very much part of the joke.
uncivilized•39m ago
I don’t know if it’s showcasing that AI is still that bad, or if it’s purposely built to be this bad in a funny type of way.
frankdenbow•1h ago
well done, got a good laugh out of this one.