Viewing profile — edg5000

And youtube shorts. Highly addictive (from a reliable source).

I bet a lot of time goes into scrolling reels, instagram, tiktok and youtube shorts. That's leisure, but not social. Of course you can send people messages and comment on things, b…

Very interesting point you make! Before LLMs I had a theory that we cannot make something more intelligent/complex than us. LLMs are certainly more knowledgable, but maybe not more…

My bank forces me to have an iOS/Android phone for online payments. So at best I'd have a cheap Android phone in a drawer in my house in case I need to make online payments. I'm su…

If you had physicial access to the machine, you could disable that feature. It would take some work since it must be baked into binaries, but far easier than creating a new EUB mac…

It probably did in China. In EU/US we don't benefit from these low prices for protectionist reasons.

Yes, that's my read as well.

Great analogy!

If we compare OpenAI subscription prices vs. the cheapest inference providers on OpenRouter, than OpenAI must be losing money. Add the fraud to this and the question is what the fu…

Can't they use GlobalFoundries, Intel or TSMC in addition to their own fabs?

Over the last 1.5 years, they had a few failures with their auth system; two or three times auth failed for a few hours so I could not work. But otherwise they have been just fine.…

I always let GPT 5.5/5.6 sol reach compatction. I'm only offered a 258K window, but I think once they release 1M to normal plan users, it will be usable across the whole 1M. At lea…

I initially thought that getting agents to work for longer and in large groups was the future, but I'm increasingly thinking that, at least for engineering, just one thread makes m…

I generally use: - Overbuilt: Designed to handle more of the same. Too strong. - Overengineered: Complexity not justified by the requirements. Sometimes overbuilding (e.g. using re…

I could not disagree more. You must have used the better models by now. Even if you only lightly use them for a question from time to time, it must have become clear that these thi…

With a coding plan, there is no need to use lower quality models right (GPT 5.6, Fable 5). Only with some kind of batch job (processing PDFs) I could imagine not having enough usag…

Yes, but you could have an uncached dynamic tail, this would not kill the cache though. And you could still periodically kill the cache, still keeping the same prefix every N reque…

It's a very interesting technology, so it leaves some shockwaves while it's impacting the world. Over time this will be absorbed and the insanity will die out, but it's quite trans…

That makes sense.

Yeah, good point. The simpler solution eventually wins. But I do think a GUI makes more sense though, that was the reason I went my own way, but I think you're right that keeping i…

Great speeds. So a small model on a lot of slow ram is fast. Now I wonder how larger models run on that thing.

I do about 10 google search queries for every 1 opus/gpt prompt. For google, I don't actually open pages anymore 9 out 10 times; I rely on the AI summary. It's fast and accurate; t…

I ask it to propose revisions, which I then review and provide detailed feedback on. It then works, and afterwards I review. I always commit beforehand. This works well for me.

I've experimented with completely novel tool call syntaxes combined with ordering and structuring the context differenly, and they seem to have a high tolerance for this stuff. But…

I spent 6 months frantically working fulltime on a harness, and when it was finished, I just stopped using it and went back to Codex CLI. I've also stopped using all the fancy feat…