zero cal pizzas, finally

Reminder that “a coin toss” is only bad odds for problems with binary and equally likely outcomes. And that’s rarely the case for anything that an LLM is used for. A 50% chance of saving an hour of work a couple times a day are pretty good odds. If I have a problem which a candidate solution is easy to verify, it’s often more effective to let an LLM investigate it for some time before I do so, and only jump in if it fails.

There have been several little fixes I’ve done in minutes with an agent that would take me at least an hour to manually investigate, write a solution, test, and refactor. So yes, there is something to it, but you need to know how to use it. Keep arguing in a thread after noticing hallucinations is a clear sign the author doesn’t know how to use an LLM.

This. If you’re at a point that you’re arguing with an LLM, you’ve already lost. Just start a new thread with a different approach, don’t make an article about your inability to use an LLM.

being annoying online

hmm

5min at 410 + 10min at 425, perfect avg

just keep rotating

That assumes Samsung, SK Hynix, and Micron won’t have competition in the next few years, but that’s already not true with the Chinese CXMT and YMTC. And the more they drive the prices up, the highest the reward for a new competitor to get established. They have a few good years (for them) charging these prices, but it won’t last.

great, let’s do it with politicians

well, you’re free to die on that hill I guess

Taking what you give it into context it’s kind of the whole selling point of LLMs. If you get garbage output maybe you’re using it wrong. In the end it’s just a tool, if you can’t use it, that’s on you.

have you seen some human written code before? Someone who knows what they’re doing with AI can definitely generate better code

If I got an AI-written codebase hard to navigate, I’d bill the AI usage it took to clean it up and, on top of it, the hours I have to put to actually do the job. We can definitely use AI to write good code, but it takes the kind of professional criteria that vibe coding lacks in order to do that.

I don’t really care about owning for its own sake, but I know services only get worse for customers over time, so that makes me prefer owning some things.

“Piracy is almost always a service problem and not a pricing problem”

yeah, the 5 most popular models in open router are chinese; and in volume of tokens, 80% of the top 10 are from chinese ones

I was reading this yesterday

https://tailscale.com/docs/reference/faq/other-vpns

I probably won’t do it myself, but maybe it works for you

that was the tutorial? Fuuck

I’m glad I’m not in the bubble of sending and receiving emails to/from robots. But to be fair, if you write emails like people in the epstein circle, maybe this is a good thing.

same, I used it for a few months, but I realized I was just saving a few dimes on every purchase, so it was barely worth the hassle. Instead I just rotate brands and dollar cost average fuel (i.e. spend the same $ amount every time instead of filling it up).

never heard of it before, that’d be so good to have worldwide

privilege elevator