Peehole

I think Claude Code was genuinely useful until like version *.0.76 or something since they bumped to *.1 it has become a shitshow. Don’t know any alternative tho

Idk what to say, that’s wild

This was a fascinating read even though I couldn’t follow a lot of stuff tbh

I did actually work on some crypto prototypes using AI and LLMs do hallucinate wallets. I was curious once because there was some wallet connected to my project so I sent like a fraction of a cent to it to see what would it happen and it got immediately drained so I checked out the wallet and I think someone’s private keys ended up in the training data. Was pretty funny to observe but it’s scary to think that people might actually lose money like that.

Sick, imagine it gets actual crypto, will it be a real wallet? Imagine they’d have money what would they use it for? Ideally, they’d start a company and actually outsource work to humans, making them essentially the bitch of a clanker and the clanker‘s constant u-turns. "You’re right, the client doesn’t need encryption for their auth endpoints. This isn’t just about security — this is about responsible user choice and not overengineering things. Good call out!“

Now that people are running openclaw I think it’s only a matter of time

I mean I’m equally worried about China and their genocide against e.g. the Uyghurs as I am about the western genocide against e.g. Palestinians or the Russian genocide against e.g. Ukrainians. The problem is with the concept of empire, not any implementation of that concept specifically.

Eh I don’t disagree with you, it’s just the reality for me that I am now expected to work on much more stuff at the same time because of AI, it’s exhausting but at least in my job I have no choice and I try to arrange myself with the situation.

I sure lost a lot of understanding of the details of the codebase but I do read every line of code these LLMs spit out and manually review all PRs for obvious bullshit. I also think code quality got worse despite me doing everything I can to keep it decent.

With proper prompting you can let it do a lot of annoying stuff like refactors reasonably well. With a very strict linter you can avoid the most stupid mistakes and shortcuts. If I work on a more complex PR it can take me a couple days to plan it correctly and the actual implementation of the correct plan will take no time at all.

I think for small bug fixes on a maintainable codebase it works, and it works for writing plans and then implementing them. But I honestly don’t know if it’s any faster than just writing the code myself, it‘s just different.

It’s kind of an interesting situation because it seems like he’s not in the Epstein files, and while I don’t believe he’s anything but a typical corrupt, opportunistic politician his opportunism is at least left-leaning and not pro-Zionism big tech.

Sadly this is a rare thing in today’s world, but also a reason why the surveillance tech apparatus hates him so I understand the urge to regulate them, but as usual this is at the expense of our freedom and just self-serving.