frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Lightweight and extensible compatibility layer between dataframe libraries

https://narwhals-dev.github.io/narwhals/
1•kermatt•56s ago•0 comments

Haskell for all: Beyond agentic coding

https://haskellforall.com/2026/02/beyond-agentic-coding
2•RebelPotato•4m ago•0 comments

Dorsey's Block cutting up to 10% of staff

https://www.reuters.com/business/dorseys-block-cutting-up-10-staff-bloomberg-news-reports-2026-02...
1•dev_tty01•7m ago•0 comments

Show HN: Freenet Lives – Real-Time Decentralized Apps at Scale [video]

https://www.youtube.com/watch?v=3SxNBz1VTE0
1•sanity•8m ago•1 comments

In the AI age, 'slow and steady' doesn't win

https://www.semafor.com/article/01/30/2026/in-the-ai-age-slow-and-steady-is-on-the-outs
1•mooreds•16m ago•1 comments

Administration won't let student deported to Honduras return

https://www.reuters.com/world/us/trump-administration-wont-let-student-deported-honduras-return-2...
1•petethomas•16m ago•0 comments

How were the NIST ECDSA curve parameters generated? (2023)

https://saweis.net/posts/nist-curve-seed-origins.html
1•mooreds•16m ago•0 comments

AI, networks and Mechanical Turks (2025)

https://www.ben-evans.com/benedictevans/2025/11/23/ai-networks-and-mechanical-turks
1•mooreds•17m ago•0 comments

Goto Considered Awesome [video]

https://www.youtube.com/watch?v=1UKVEUGEk6Y
1•linkdd•19m ago•0 comments

Show HN: I Built a Free AI LinkedIn Carousel Generator

https://carousel-ai.intellisell.ai/
1•troyethaniel•21m ago•0 comments

Implementing Auto Tiling with Just 5 Tiles

https://www.kyledunbar.dev/2026/02/05/Implementing-auto-tiling-with-just-5-tiles.html
1•todsacerdoti•22m ago•0 comments

Open Challange (Get all Universities involved

https://x.com/i/grok/share/3513b9001b8445e49e4795c93bcb1855
1•rwilliamspbgops•23m ago•0 comments

Apple Tried to Tamper Proof AirTag 2 Speakers – I Broke It [video]

https://www.youtube.com/watch?v=QLK6ixQpQsQ
2•gnabgib•24m ago•0 comments

Show HN: Isolating AI-generated code from human code | Vibe as a Code

https://www.npmjs.com/package/@gace/vaac
1•bstrama•26m ago•0 comments

Show HN: More beautiful and usable Hacker News

https://twitter.com/shivamhwp/status/2020125417995436090
3•shivamhwp•26m ago•0 comments

Toledo Derailment Rescue [video]

https://www.youtube.com/watch?v=wPHh5yHxkfU
1•samsolomon•28m ago•0 comments

War Department Cuts Ties with Harvard University

https://www.war.gov/News/News-Stories/Article/Article/4399812/war-department-cuts-ties-with-harva...
6•geox•32m ago•0 comments

Show HN: LocalGPT – A local-first AI assistant in Rust with persistent memory

https://github.com/localgpt-app/localgpt
1•yi_wang•33m ago•0 comments

A Bid-Based NFT Advertising Grid

https://bidsabillion.com/
1•chainbuilder•36m ago•1 comments

AI readability score for your documentation

https://docsalot.dev/tools/docsagent-score
1•fazkan•44m ago•0 comments

NASA Study: Non-Biologic Processes Don't Explain Mars Organics

https://science.nasa.gov/blogs/science-news/2026/02/06/nasa-study-non-biologic-processes-dont-ful...
2•bediger4000•47m ago•2 comments

I inhaled traffic fumes to find out where air pollution goes in my body

https://www.bbc.com/news/articles/c74w48d8epgo
2•dabinat•47m ago•0 comments

X said it would give $1M to a user who had previously shared racist posts

https://www.nbcnews.com/tech/internet/x-pays-1-million-prize-creator-history-racist-posts-rcna257768
6•doener•50m ago•1 comments

155M US land parcel boundaries

https://www.kaggle.com/datasets/landrecordsus/us-parcel-layer
2•tjwebbnorfolk•54m ago•0 comments

Private Inference

https://confer.to/blog/2026/01/private-inference/
2•jbegley•58m ago•1 comments

Font Rendering from First Principles

https://mccloskeybr.com/articles/font_rendering.html
1•krapp•1h ago•0 comments

Show HN: Seedance 2.0 AI video generator for creators and ecommerce

https://seedance-2.net
1•dallen97•1h ago•0 comments

Wally: A fun, reliable voice assistant in the shape of a penguin

https://github.com/JLW-7/Wally
2•PaulHoule•1h ago•0 comments

Rewriting Pycparser with the Help of an LLM

https://eli.thegreenplace.net/2026/rewriting-pycparser-with-the-help-of-an-llm/
2•y1n0•1h ago•0 comments

Lobsters Vibecoding Challenge

https://gist.github.com/MostAwesomeDude/bb8cbfd005a33f5dd262d1f20a63a693
2•tolerance•1h ago•0 comments
Open in hackernews

Ask HN: From the MIT study, is it smarter to resign than to use forced AI?

12•ciwolex•7mo ago
Below is the link to the MIT study for reference.

https://arxiv.org/pdf/2506.08872

Comments

pavel_lishin•7mo ago
I don't think a lot of us here write essays for a living.
savorypiano•7mo ago
This paper makes me glad I am not a researcher.
AnotherGoodName•7mo ago
Shrug If you are being told by your boss to use AI or quit and you absolutely refuse to use AI yes you should quit.

The end. No anger at you or your boss. It's an incompatibility.

I understand why some companies mandate usage these days. Especially for programming. The honest truth is that it does speed up development. The other honest truth is that there's resistance to change that harms productivity at times like this and the only way around that is for leadership to be very direct on this point.

To use a metaphor employers don't want weavers who refuse to make use of the loom. https://en.wikipedia.org/wiki/Luddite

p0w3n3d•7mo ago
The loom however does not make t-shirts despite you asking it for the shorts. The loom does not confabulate, create four handed sleeves and tell you this is the correct way...
techpineapple•7mo ago
Just like, thinking about how bad early machines were, I bet early looms made a whole bunch of annoying mistakes that required cleanup. I think one of the biggest jobs of the early Industrial Revolution was like babysitting machines.
HPsquared•7mo ago
Most technical jobs today are still babysitting machines, in some form.
bgwalter•7mo ago
You are a good employee!

> To use a metaphor employers don't want weavers who refuse to make use of the loom.

The loom actually produced something, as opposed to mediocre coders who ingratiate themselves with management by pushing "AI" because they hate all productive people.

The loom also did not steal other people's IP.

AnotherGoodName•7mo ago
I see this attitude a lot. The idea that people finding value from AI are mediocre programmers.

I've been staff level at Meta and Google and many other companies in my career. I've been in the industry over 20 years now. I can talk to peers in the industry at that level and above and the sentiment is pretty universal. "This saves a lot of time, we need engineers to learn to use this asap". Such decisions are not coming from a vacuum. It's literally your most senior engineers advising management that leads to these mandates.

reseasonable•7mo ago
It’s a skill just like any other. Without practice people will absolutely waste time. But knowing which problems are appropriate for an LLM can save days of labor.

As an engineer in the field for 30 years, this tech beats every man page, sdk, google search, stack overflow post or head banging situation by light years.

The rebuke of it is surprisingly antithetical to the progress of the industry. It reminds me of the decade+ of those who refused to use an IDE and favored notepad[++] or vanilla vim instead. I suspect in 5 more years those similarly-minded folks will finally adapt in the same way we’ve all grown accustomed to IDEs today. Or those who adapt will just outpace those who don’t.

happytoexplain•7mo ago
The rebuking is largely focused on using AI-written code (and other content) in ignorant ways, not on using AI to learn or to get example code upon which to base your own, or for simple objective code you can confidently audit, which are in fact more reliably good and less depressing/degrading/market-damaging use cases.
tonfa•7mo ago
> To use a metaphor employers don't want weavers who refuse to make use of the loom

it's funny to reference that, since the luddite movement was about working conditions, pay, and quality of goods produced. It wasn't ideological opposition to technology (They didn't destroy machines when acceptable conditions were agreed)

happytoexplain•7mo ago
>No anger at you or your boss. It's an incompatibility.

I am allowed to be angry at a person for their opinions or actions.

>Luddite metaphor

Metaphors are already a weak and oversimplifying rhetorical hammer, this one particularly so.

frizlab•7mo ago
> The honest truth is that it does speed up development.

That is a highly controversial statement! It depends on what you’re working on, how you work, and so many other factors than just laying it out as just a fact is close to dishonest.

brudgers•7mo ago
Smarter is a very poor metric.

And a convenient excuse.

BoorishBears•7mo ago
Past a certain level of seniority, your job increasingly becomes translating imprecise mandates from on high into practical outcomes.

If I heard someone making a blanket mandate for using AI, I'd translate that to them hearing this "new AI thing" allows people to do more work more efficently, and they to see that increase in their org.

I'd take that mandate as a chance to explore AI on someone else's dime, but continue doing my work otherwise, only using AI as it benefits me.

If expectations rise to un-reasonable level because of unrealistic expectations around AI, that's a seperate problem you'll have to deal with.

(It's also not a great look that your leadership wouldn't dig in to realize how silly forced AI is, but a charitable reading is that they're trying to force interactions with AI so employees can discover where it works and where it doesn't)

gebdev•7mo ago
This is an interesting study. I wonder how the LLM option might compare to human written responses in the same format (but with higher latency), or even to having a physical human in the room. Given some of the points from the conclusion about teachers being able to detect LLM inspired work, I wonder if either of these options may, at times, be better forms of learning due to improved quality.
m3047•7mo ago
+1 for the article, don't have an answer to your question.
deepsummer•7mo ago
Let me ask you a related question: if there was a study that handwriting is better for your brain than typing, should secretaries have quitted when typewriters and computers were introduced?

The thing is, there is no going back. There will be no significant demand for output that's created by humans even though a machine can do it as well. You can try to find a niche where AI is worse than humans. But that will be increasingly difficult to find.

So if you want to continue doing things without AI, that's fine. But most likely it will be a hobby, not a job.