someoneontenet

Curriculum learning helped me out a lot in this project too https://www.robw.fyi/2025/12/28/solve-hi-q-with-alphazero-an...

How did I know this was gonna be speed lol

Love these write ups!

Come visit for a few weeks and see if it’s right for you :)

I’ve tried to set this up and get a little bogged down on some of the finer points of compiling nixos to wasm. It’s not very straightforward unfortunately :/

My dream is have a in browser nixos vm on wasm. If I could have a bare vm, I can bootstrap it easily with a nixos config. From there I can start thinking about running web services in browser tabs instead of physical hardware.

I think that you might be taking that example in bad faith.

At that point you may as well use a typewriter

Neat, I’ve also made something like this with processing https://robw.fyi/constellation.html

The readme was the most important part for me!

Sounds like op is manually fuzz testing.

I talk to ai like a caveman mostly. Instead of over optimizing my prompt I just try to find the minimal amount of representation to get the llm to understand my problem and solve it for me and I have been very productive with this strategy. What would someone like me get out of prompt engineering? Or is it more for things like agents, bots, and assistants?

Seems weird to compare them since I see them as complimentary services.

I’m sure some countries are more willing to treat guests to the vegetable treatment than others.

How is this done for js?

Fantasize?

Newtons Method approximates square roots. Its useful if you want to approximate something like that without pulling in the computational power required of NN.