News for hackers. Stuff that matters.

ex-aws-dude

We just don't do that, I don't know why people talk like that is a given

Is it a company culture thing or a domain specific thing?

Ask HN: How can we stop UX people from changing UIs to get promotions?

Every product seems to go through constant UI churn that has no benefit to users other than changing around whitespace or refreshing the look.If you have a full time UX person they _must_ change the UI of your product otherwise there is no way to get promoted.That is just an unavoidable fact of incentive structures.As a company how can you avoid that issue?

( Read More... | 6 comments | 3 points | on HN )
Re:Tesla Roadster on Friday October 16, 2026 @05:59PM (#50125307)
What is demo'd and what actually ships are two completely different things

I'm not saying its impossible but it seems pretty naive to just assume on face value

Re:Why is every rationalist blog post way too long? on Thursday October 15, 2026 @01:09AM (#50099418)
It’s because the point isn’t actually to convey the idea

It’s to do the practice of rationalist blogging

Re:Ask HN: Many SWEngs out there are considering re-training on Wednesday October 14, 2026 @02:06PM (#50092180)
IMO its better to just wait and see how it shakes out rather than overreacting

Think about people who left the industry in the dot-com era and how that probably seemed smart at the time

Ask HN: Are you hesitant to install vibe-coded apps?

The thing is in the past there can be supply chain attacks but when its an actual programmer creating the software there is at least a chance for them to notice if something is up.But if the app is not even created by a programmer whats stopping prompt injection from including some virus in the app?Aren't we completely blind to security issues at that point?

( Read More... | 25 comments | 9 points | on HN )
Re:Ask HN: How do you review code nowadays? on Tuesday October 13, 2026 @03:39PM (#50077449)
I download the change into my local context and use AI to give me a high level overview of the change because people always sucked at describing their changes anyway

Then I go through a review it as normal and flag issues

Then I do a pass asking my own local agent to find any issues, I sanity check/confirm those and comment them if necessary

I'm talking about defrosting or shutting off the fridge specifically

What use case does that need to be possible over the internet?

Re:Redesigning Source Control at Anthropic on Tuesday October 13, 2026 @03:36PM (#50077398)
It is really laughable when you zoom out and realize that nothing much has really changed day to day
Re:Apple HomeView on Tuesday October 13, 2026 @12:04PM (#50073895)
Guys apple didn't invent this form factor, Alexa had Echo Show 10 at least 5 years ago and alexa with a screen way before that
Re:Leaving Tech on Monday October 12, 2026 @11:06PM (#50066097)
That’s a bit harsh but it does sound like they aren’t willing any compromise which is a bit unreasonable in the current market

Also 36 jobs isn’t that much

Why is that even internet enabled?

If you want to defrost it presumably someone already needs to be physically present to unload the fridge?

Re:Ask HN: Why aren't Americans rising up against tipping? on Saturday October 10, 2026 @11:08PM (#50039466)
Thats totally unrelated

I’m saying people are willing to accept marginal worseness the same way the frog in a pot of water accepts an increase of 1 degree

Re:Ask HN: Why aren't Americans rising up against tipping? on Saturday October 10, 2026 @09:03PM (#50038740)
Because it’s only marginally more expensive
Re:Reactions to 100 Solutions on Saturday October 10, 2026 @09:07AM (#50032570)
It feels like I’m reading a historical document from the industrial revolution
Re:OpenAI Withdraws 3 Math Papers on Friday October 09, 2026 @09:38AM (#50020387)
Exactly, big O doesn’t claim it’s actually faster in practice on a computer for all inputs
If you can’t even take the time to understand what you’re clicking why even go through the formality of “approving”
Re:The Mathocalypse on Thursday October 08, 2026 @08:05PM (#50014169)
It does make you ask: what’s more likely
Re:Sharing AI progress in mathematics on Tuesday October 06, 2026 @11:52PM (#49987926)
Wouldn’t that just result in wireheading
Re:Sharing AI progress in mathematics on Tuesday October 06, 2026 @10:05PM (#49987103)
The scoop gun
Re:Sharing AI progress in mathematics on Tuesday October 06, 2026 @07:50PM (#49985865)
That’s always been a thing, it’s called “getting scooped”
wouldn’t it result in whoever has the most money having the most secure software?
Hey man I'm just looking up the symbols like they told me
Well the whole lesson learned was that the Turing Test as it was defined was way too easy, it was a bad criteria for GI because it underestimates how easily humans find meaning/patterns in things.

I mean you could show people random markov chain gibberish in 1996 and they would swear they found intelligent meaning in it

https://en.wikipedia.org/wiki/Markovian_Parallax_Denigrate

its not, from what I've heard the difference in quality was not worth it
Re:Ask HN: What're You Reading? on Thursday October 01, 2026 @02:29AM (#49918367)
I finished it and thought it was overall good but I don’t know if I would say it’s good “bang for buck” in terms of what you get for the time commitment
Re:Gemini 4 Argon on Thursday October 01, 2026 @12:14AM (#49917668)
Because you don’t need TPUs and anyone can scrape the internet

I think that has been well proven so far

Re:LinkedIn Larpmaxxing on Wednesday September 30, 2026 @01:40AM (#49904789)
Working games mine is all layoff posts recently

Just scrolled now and saw like 5

Re:Ask HN: What're You Reading? on Tuesday September 29, 2026 @05:03PM (#49900511)
Reading Gödel, Escher, Bach but I've been stuck on the MU puzzle for days
Re:I was impressed by Jev, please explain why I shouldn't be on Monday September 28, 2026 @05:58PM (#49885031)
> It's doing something more novel than the haters suggest

How is that different than https://www.nobodywho.ai/posts/jev-in-25-lines/

Couldn't you always get probabilities for a fixed set of tokens without training a new model?

Older »