This is awful, of course.
It seems to me it does not matter how many cores your processor has or how much RAM you have. My machine with 128 gigs of RAM still manages to crawl.
I have Claude, Codex and Cursor running, I think. And these guys manage to hang really hard.
Yes, I probably have 40 sessions running in parallel by now. But, f$ck, folks, what is going on here, this is a sh$tshow.
I have started thinking I need to buy some kind of server.
The only thing saving me is that there are five computers around me. That is, while one computer is crawling, I move over to another. Because it is impossible to hold everything on one machine.
Honestly, I thought I had a pretty powerful computer. But now I have realised I am short on cores. And the RAM needs to be not 128 gigs any more but maybe 256.
In short, some kind of trash, honestly.
And the thing is, my LLMs are not local. Everything of mine is in the cloud, all these LLMs are cloud ones. It is just Python scripts executing.
And no, of course, a lot of things are not optimised over there. I have the syncing, the flows, well, nothing.
But trash. Awful.
The full story, in two versions:
📖 For humans, the longread: https://github.com/tonydzi/clawrush/blob/main/longreads/20261022.md
🤖 For machines, the devlog: https://github.com/tonydzi/clawrush/blob/main/devlog/20261022.md. Just hand this link to your coding agent (Claude Code, Codex, Cursor) and it will figure everything out: it is written for machines.
🔗 All our channels and contacts in one place: https://linktr.ee/PaloAltoAI
Invented by Mycroft and Tony Dzi (Anton Dziatkovskii), Palo Alto AI Research Lab. Proudly made in Silicon Valley.
Top comments (0)