frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Native-speed vLLM transformers modeling back end

https://huggingface.co/blog/native-speed-vllm-transformers-backend
1•simonpure•35s ago•0 comments

Google Vibe Coding vs. Agentic Engineering White Paper

https://www.kaggle.com/whitepaper-the-new-SDLC-with-vibe-coding
1•grepsedawk•57s ago•0 comments

A tool that checks how your business appears in AI search

https://www.tryagentscore.com
1•nikhilsiyer•2m ago•0 comments

Every Way Meta Tracks You, and How to Stop It [video]

https://www.youtube.com/watch?v=Yv2Eb_kJous
1•nalekberov•5m ago•0 comments

DSS Code Prime

https://github.com/dailysoftwaresystems/dss-code-prime
2•rgasperetti•6m ago•0 comments

Are Argentina being treated favourably at World Cup?

https://www.bbc.co.uk/sport/football/articles/cx2wkwd7e6go
1•password54321•7m ago•0 comments

Show HN: Day planner shaped like a clock (2-way calendar and Todoist sync)

https://reassign.app
1•smuk3c•8m ago•0 comments

Adsb.fi

https://adsb.fi/
1•rolph•8m ago•0 comments

Lessons from the Vasa Shipwreck

https://www.ft.com/content/200a6c44-9b66-4af3-82eb-98acb53898e4
1•bookofjoe•10m ago•1 comments

'Acceleration without fuel:' superconducting thruster in first orbital test

https://www.space.com/technology/acceleration-without-fuel-revolutionary-superconducting-thruster...
1•breve•13m ago•0 comments

Remote sync, MCP, and an API for your Obsidian vault

https://github.com/8thpark/geode
2•pmihaylov•15m ago•0 comments

The neutral proof standard for consequential AI-agent actions

https://github.com/Actenon
1•Bucko1•18m ago•0 comments

How much and why ACA Marketplace premiums are going up in 2027

https://www.healthsystemtracker.org/brief/how-much-and-why-aca-marketplace-premiums-are-going-up-...
2•LostMyLogin•18m ago•0 comments

A Treatise on How to Use the Internet Without Committing Philosophical Suicide

https://pastebin.com/Ft7P5m9F
1•jxmorris12•23m ago•0 comments

North Korean Hackers Compromise Go and PHP Packages

https://opensourcemalware.com/blog/polinrider-jumps-the-fence
2•6mile•27m ago•1 comments

Rewriting Bun in Rust

https://bun.com/blog/bun-in-rust
30•afturner•27m ago•2 comments

Show HN: Screenstab (tilt-shift-style screenshots) is now free and open source

2•mikaelaast•29m ago•0 comments

CrowdSound: A Song Composed Anonymously by the Internet

https://web.archive.org/web/20191208165423/https://crowdsound.net/lyrics
2•Jomal_HN•32m ago•0 comments

Open Source Barware: free, local-first bar inventory software (GPLv3)

https://opensourcebarware.com
2•RichBJamison•34m ago•0 comments

Apple's AWDL causing periodic 90ms ping on LAN

https://twitter.com/tomaskafka/status/2074963014596366670
2•tomaskafka•36m ago•0 comments

Build Your World Cup Dream Team

https://7-0-game.com/
1•jeyzolo•37m ago•1 comments

More Workers Take Mental Health Leave, and Bosses Aren't Happy

https://www.bloomberg.com/news/articles/2026-07-08/mental-health-leave-is-rising-as-more-us-worke...
6•wslh•37m ago•2 comments

BitTorrent's disastrous, legendary, and controversial story

https://www.theverge.com/tech/959848/bittorrent-story-25-years-piracy
3•colinprince•39m ago•0 comments

Robust Secret Storage in Networks

https://arxiv.org/abs/2606.30261
1•Anon84•39m ago•0 comments

Show HN: Probed – Talk to your People. Customer chat, feedback, and roadmaps

https://probed.chat/
1•HeadOfProbing•39m ago•0 comments

Lucky young couple lands gig taking care of uninhabited Irish island

https://www.cbc.ca/lite/story/1.7477696
3•colinprince•40m ago•0 comments

Reframing smart glasses as 'pervert glasses'

https://this.weekinsecurity.com/reframing-smart-glasses-as-pervert-glasses/
8•g-b-r•40m ago•2 comments

Does the Recent SCOTUS Geofence Case Threaten Flock?

https://www.youtube.com/watch?v=R_fVvppq7rg
3•sbuttgereit•42m ago•1 comments

Shares of gun seller GrabAGun, backed by Don Trump Jr, tank on NYSE debut (2025)

https://www.cnbc.com/2025/07/16/trump-jr-grabagun-stock-gun-merger.html
5•xrd•44m ago•0 comments

Benchmarking Coding Agents on Databricks' Multi-Million Line Codebase

https://www.databricks.com/blog/benchmarking-coding-agents-databricks-multi-million-line-codebase
1•tanelpoder•47m ago•0 comments
Open in hackernews

Separating signal from noise in coding evaluations

https://openai.com/index/separating-signal-from-noise-coding-evaluations/
79•sk4rekr0w•1h ago

Comments

2001zhaozhao•1h ago
Translation: other labs have learned to benchmaxx SWE-Bench Pro better than they do
xacky•1h ago
Achieving AGI will be more than just passing all benchmarks, it has to account for the unknown problems too.
metalliqaz•41m ago
Unless they have something in the labs that massively departs from their current products, AGI isn't on the table and is purely hype for marketing purposes.
cyanydeez•33m ago
they should be consulting Donald Rumsfeld and make sure they implement the Unknown-Unknowns benchmark, because thats how they get you
minimaxir•33m ago
This ties into the bias-variance tradeoff (https://en.wikipedia.org/wiki/Bias%E2%80%93variance_tradeoff) common with building non-LLM models. The solutions can only be a) figure out how to get LLMs smaller with similar performance so they don't memorize things/game the benchmarks and b) build benchmarks that are indeed comprehensive for all real-world data, which is infeasible.
sigbottle•21m ago
I mean, people always say there are tradeoffs, until you reach the next frontier, in which there are tradeoffs at said frontier, and the next, and the next, etc.

In one sense, yes, tradeoffs are inescapable as the scope expands to the maximal possible scope. In another sense... it depends on the level of abstraction we're talking about.

naikrovek•28m ago
AGI is a long way off. Unless you’re talking about some unknown-to-me LLM marketing BS which is called “AGI” or something, I guess. Artificial general purpose intelligence is so different to LLMs or image AI that they are completely incomparable, except to say that they are all artificial. AGI will do a lot more than token prediction.
bellowsgulch•58m ago
Seems like depending on your field these days, the hot thing to do is build your own private benchmarks.

In my own testing, no frontier model knows how to replicate an original 1990s Super Soaker prototype design, which for the most part, should be almost completely possible with Home Depot parts.

They just don't understand PVC parts, triggers, etc.

softwaredoug•56m ago
Or defensively expect models to be stupid.

Seems the smart thing to do is not assume an agent will do the right thing. But to create the scaffold / harness that enforces constraints to steer them towards a good result.

Then you can swap out the really smart model for maybe something cheaper.

thierrydamiba•51m ago
Or you’re getting steered into la la land because of your prompt
bellowsgulch•48m ago
Certainly, but deconstructing the problem, none of the models seem to appreciate the staggering difference between a ball valve and a button release.

Of course, there's also no super soaker engineer jobs to take, so I'm sure training sophisticated models to do well in that area is not a high priority for any firms.

ACCount37•18m ago
It's a "commonsense spatial reasoning/problem solving" kind of problem. LLMs fail at spatial reasoning forever.

What humans "easily" solve in seconds with raw spatial reasoning LLMs often find easier to solve by invoking A* or a constraint solver.

Might be that text data is particularly bad at teaching that to LLMs. Or it might be that human brain was optimized by evolution for solving spatial problems in open ended 3D environments for hundreds of millions of years, and only optimized for writing computer code for a few decades at most.

The current frontier is halfway competent at benign closed 2D work, but still completely fumbles anything remotely close to open ended real world 3D work.

ReptileMan•50m ago
Lately my benchmark is build123! - trying to force them to build me functional parts only by the description. All of the models don't perform well.
mgiampapa•43m ago
IDK, sounds like it has brute forced my password already.
midtake•29m ago
This guy builds
shay_ker•47m ago
Didn't we all know from the start that all of SWE-Bench was flawed? Even the authors concede the limitations and have long since moved on.
paxys•41m ago
SWE-Bench Pro was created to replace SWE-Bench and fix these problems.
warkdarrior•36m ago
SWE-bench Verified was created to fix the problems of SWE-bench.

Then SWE-Bench Pro was created because SWE-bench Verified had flaws.

Now SWE-Bench Pro is shown to have flaws.

carabiner•30m ago
Is there a way to benchmark the accuracy, validity improvements in these successive benchmarks?
jaggederest•6m ago
Bench Bench Pro Maxx Series S 360? The original Bench Bench Pro Maxx Series S had some quality issues, so that's the current followup. We've also released a higher order benchmark developed out of Bench Bench Pro Maxx Series S 360 One King Ranch edition, allowing future benchmark towers to be fully self-contained.
denysvitali•26m ago
Well, we now have DeepSWE
dandaka•45m ago
What is considered SOTA for SWE benchmarks now?
EuanReid•43m ago
I've generally found DeepSWE[0] to be pretty true to reality.

[0]: https://deepswe.datacurve.ai/

enraged_camel•42m ago
FrontierBench
dandaka•24m ago
do they have a website? I have found only paper PDF and it seems more general than SWE
carabiner•41m ago
strawberry
retr0rocket•32m ago
Why is this a problem? Its like asking a person how many elder futhark runes are in the word strawberry.

Unless you want to tack on bpe enconding table to every llm context its pointless

swyx•30m ago
https://cognition.ai/blog/frontier-code (disclaimer - was on the team - but also we covered swebench pro/deepswe issues in here as well.)
porphyra•40m ago
Interesting timing to release this just when SWE-1.7 and Grok 4.5 came out being much cheaper than GPT-5.5.
johngoode•33m ago
This doesn’t seem like opportune timing to announce days before a new model drop
janalsncm•21m ago
Based on the numbers here it seems there’s less than 800 tasks in the entire benchmark. That is enough for a handful of engineers to comb through in a week (which is what OpenAI eventually did here).

On the one hand, kudos to them for actually doing that work.

On the other hand, garbage in, garbage out. It’s a bit embarrassing for the original authors to have not actually checked, and it’s embarrassing for everyone downstream to have not checked either.

Also if you check the article, although an LLM did find issues, it tended to underestimate issues that professional software engineers found.

jheitmann•20m ago
It reads to me like "We did all the work you'd do to figure out how to fix the benchmark, then we decided to throw out the benchmark". Is there some reason the underlying data is so golden that it can't be patched? At the end they argue for a slightly more curated approach to benchmark generation, but my gut is that using messy ill-specified tests taken from real world data and patching them into fairness would be a pretty solid path to take.
Centigonal•14m ago
If they fixed it, then it wouldn't be SWE-Bench Pro anymore, right? It'd be "SWE-Bench-Pro-Fixed-OpenAI." I think it's better optics for the independence of the benchmark if the OpenAI team lets some third party do the fixing and release the improved benchmark.

...Although OpenAI did exactly that when they released SWE-Bench Verified, so maybe I'm talking out of my butt here.

tedsanders•10m ago
Pointing out problems (e.g., hidden tests that assume narrow implementation details) is harder than fixing problems (e.g., creating tests that work for any possible choice of implementation).
mlhpdx•13m ago
Fundamentally aren’t they concluding that tasks assigned to software developers (human or otherwise) are often incomplete, self contradictory or worse? This is the world in which their tool must play. I’m unsympathetic.
CSMastermind•10m ago
DeepSWE is the one I generally trust: https://deepswe.datacurve.ai/
Topfi•29m ago
Either DeepSWE [0] or FrontierCode [1], depending on personal goals and requirements. The later is more interesting for me personally, due to the design of the benchmark heavily grading "mergability", i.e. how the provided output is to review and whether a serious developer can easily parse it and'd be willing to merge the result. In my mind and with my private evals, for quite some time I've held firm that a model can have a higher ceiling but that has limited value if I do not feel truly confident in signing off on the code.

[0] https://deepswe.datacurve.ai/

[1] https://cognition.com/blog/frontier-code-1.1