Hacker Newsnew | past | comments | ask | show | jobs | submit | rafram's commentslogin

Its context includes reasoning that you can’t see, so this is actually a reasonable thing to ask.

> Opus or GPT-5.5 are the only ways to even attempt this.

It’s pretty funny to claim that a model released 22 hours ago is the bare minimum requirement for AI-assisted programming. Of course the newest models are best at writing code, but GPT-* and Claude have written pretty decent systems for six months or so, and they’ve been good at individual snippets/edits for years.


> It’s pretty funny to claim that a model released 22 hours ago is the bare minimum requirement for AI-assisted programming.

Not what I said.

The OP was trying to write specs and have an AI turn it into an app, then getting frustrated with the amount of cleanup.

If you want the AI to write code for you and minimize your cleanup work, you have to use the latest models available.

They won't be perfect, but they're going to produce better results than using second-tier models.


Is it actually the case that 5.5 is that much better at implementing specs than its very capable predecessor released a month ago? Just seems like a baseless and silly claim about a model that has barely been out long enough for anyone to do serious work with it.

> Is it actually the case that 5.5 is that much better at implementing specs than its very capable predecessor released a month ago?

The OP comment was talking about Claude Sonnet. I was comparing to that.

I should have just said "use the best model available"


> Is it actually the case that 5.5 is that much better

Nobody was talking about how much better it is until you wrote this though

It's like you're building your own windmills brick by brick


An image having a bit of bloom doesn't make it AI generated. If the Times started using undisclosed AI-generated images, it would be a huge scandal, and a violation of their own policies: https://www.nytimes.com/2025/06/12/reader-center/new-york-ti...

I think this should've been two separate blog posts.

Yeah, it’s funny how all the comments so far are only talking about the over-engineering and scope creep, when the bulk of the blog was dedicated to a totally separate rant (but a good one!) on structural diffing.

Kind of hilarious though that it talks about scope creep and then transitions into a whole different long topic.

Looks like this was a newsletter by the author, not a blogpost

That makes more sense!

Palantir develops database software.

… As part of an explicit, openly stated mission to reshape the global political order.

Palantir is indeed in many ways just a software vendor but we shouldn’t downplay that they have a much more explicit agenda than most other companies do in seeking government contracts.


Eh. I mean, the government will do what the government will do with the software it buys. We've just seen that with Anthropic. The US government wouldn't give contracts to Palantir if it seemed like its ideology didn't line up with US aims, and they wouldn't give contracts to other vendors if it seemed like their less ideological marketing meant they weren't aligned with US aims.

1) we’re talking about a UK government contract with Palantir

2) actually historically, and aspirationally, the US government isn’t supposed to be focused on ideological alignment of its vendors - the current government is anomalous and we shouldn’t normalize this.


“Palantir is here to disrupt and make the institutions we partner with the very best in the world and, when it’s necessary, to scare enemies and on occasion kill them,” Karp said, with a smile on his face. The CEO added that he was very proud of the work his firm is doing and that he felt it was good for America. “I’m very happy to have you along for the journey,” he said. “We are crushing it. We are dedicating our company to the service of the West, and the United States of America, and we’re super-proud of the role we play, especially in places we can’t talk about.” [1]

[1] https://gizmodo.com/palantirs-billionaire-ceo-just-cant-stop...


Yes, that’s a bunch of bluster about database software.

No, Palantir is not a "database vendor", it's an intelligence company closely working with IOF in their ongoing genocidal efforts and with DHS with mass deportations.

I'd rather see Oracle than a ghoul openly supporting targeting civilians.


Doesn't Oracle (or at least Larry Ellison) openly support extermination of civilians too?

Not ordinarily, at least not anymore. They cancelled Project Beanstalk in the late 2010s, now relying on the legal system to extract perceived debts.

Never to that extent FWIR but I see your point. Yeah, Oracle is bad too.

The father out from the center of the galaxy they look, the younger the stars are.

Check that you’re running the latest version.

Lots of bodega cats are allowed to go out on the street. They usually don't wander far. Cats know where home is.

> Show up like a human.

Ironic advice considering that this post seems to be fully AI-generated, and the whole thing is an ad for some vibe-coded dead man's switch subscription.


I felt that it was AI-assisted rather, and had actually decent info.

Go read what Gruber says about Trump on his social media, or even on the very blog you were just on. Safe to say he isn’t a fan. I think what he’s saying is that Cook has been quite effective at stroking Trump’s ego enough that the admin leaves Apple alone, which is absolutely true in my opinion.

In a different world where Cook messed up, it might be Apple (a Big Tech company with uber-liberal employees, marketing, and vibes, and an openly gay CEO!) being designated a supply chain risk, not Anthropic.


Consider applying for YC's Summer 2026 batch! Applications are open till May 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: