Ewww

What's bad about it and what's a better one?

AFAIK pi's approach is to be quite minimal and allow extensions for customization, making it a more flexible solution, but you need to do work to make it fit your use case. OP mentions one extension, but perhaps it'd have benefited from more.

Another choice would be opencode which has more functionality and is a more heavyweight option out of the box.

BYD cars in China are not very good. But to export them they need to be higher quality to meet criteria. I would never buy one in China. Happily own one outside of China.

No. The author is wrong. If you’re still using single model/context then it’s kinda your own fault for using things poorly.

Had a similar issue and just pointed codex at my laptop and it got everything working.

Same. By the time I got back it had even upstreamed a few device drivers to mainline Linux, all on its own.

how'd you do that when wifi and USB wasn't working?

That’s all it is. This only exists to lock you into cloudflare even more.

Then why did they deliberately make it compatible with Open Feature, explicitly making it easy to swap out a different Open Feature provider?

Oh, that's right, you just spouted a "big company bad" mantra without bothering to read the article. Look, I know saying RTFA goes against the HN guidelines, but the amount of increasingly lazy spew i see from folks (or bots) who haven't bothered to read the article is so tiresome and annoying.

It’s weird how insecure people are on HN that they need to downvote and flag comments when their feelings are upset. Instead of learning and accepting that they are wrong and there is room for improvement. They close their eyes plug their eyes and scream until the comments are flagged.

Hahahha no. lol just no.

> On civil engineering projects, I’m pretty sure a human reviews each weld.

Nope. It’s sampled.

Yeah because they are not auto regressively generated!

As a counterexample, I've been seeing more "safety rejections" from Claude. Unlike search, being unable to ask _anything_ about botulinum, or details about the recent Copy Fail vulnerability (without giving my fingerprints to Anthropic to become a "verified security researcher") we're only just beginning to see the ways LLM can be used to distort information and its availability.

That's fine if we aren't destroying existing products to replace them with AI.

People can already use AI mode in google search if they want. "It'll be better later" is a shit reason to kill one product for it.

So you started with ‘highly doubtful’ as a comment, got given lots of examples and instead of assimilating that info you closed your eyes put your fingers in yours ears and said “oh who cares?’ - you’re on team AI regardless eh? That’s fucking weird mate.

Na. Wasn’t given any good examples. People just whining about the same stuff because “oh no I got information that’s former in the same structure that I can tell it’s AI and it makes me feel bad”

My grandfather was one of the first people in Canada to own a commercially available chainsaw.

Let me tell you - it didn’t take 30 years for people to figure out that chainsaws were useful.

Yet most of the kids on HN think they write better code than ai and that it’s completely useless and has no place.

LLMs do write code better than many engineers. It’s just the the normal distribution and the way LLMs are built. Just because it doesn’t write better code than above average engineers doesn’t make it useless though.

People that say AI writes better code than them is such a tell but not the one they might think it is.

Whether or not it is worth the cost or second order effects on society is a different discussion.