The entire environment requires only 300 lines of C.
Many users dismissed the six-pendulum cartpole experiment as a useless waste of GPUs and engagement farming, while one noted commodity cloud pricing now enables advanced RL work.
Want to solve cool problems with RL? Take an afternoon to read our docs (1 page) on http://puffer.ai and build a ~300 line env. It's C but it's really easy C. Extensive guides on my articles tab for learning more from there! All free + OSS, and I review PRs on stream!
behold. THE WORLDS FIRST SIX PENDULUM CARTPOLE SOLVE. Including a sponsor!
To solve this task, I built an environment to train an AI. This is what mechanize does, but for larger AIs. Apply! Salaries are up on their page
Thank you to mechanize for sponsoring!

@airkatakana he was broke. and then roon followed him

@airkatakana 4090 is GPU type, not the number of GPUs

@devnullest i wish such a life-changing event would happen to me

@importdhruv @airkatakana

@airkatakana You’ll never believe this but back in the day we had 8800 GPUs

@airkatakana Elonbux

@airkatakana I am having a 6000 GPUs

@airkatakana it's only about 3090 after vat tax

@airkatakana lmfao

@airkatakana That's an oddly specific number of GPU's

@airkatakana what an unnecessary waste of bare metal tbqh

@airkatakana Seriously? Isn't 4090 the type of the GPU, like Pro 6000?

@fareesh @airkatakana I got it but the way it was written I felt it was genuine doubt

@airkatakana maybe he isn't actually unemployed? aws gives him unlimited gpu like openai gives steinberg

@airkatakana I’m so glad balancing 7 pendulums is the groundbreaking problem being focused on…
I do not participate in society either!

@a4dex @airkatakana how

@devnullest @airkatakana And now he has 4090 GPUs

@airkatakana Next time he will drop an OSS model twice the size of Deepseek

@airkatakana good bait
The entire environment requires only 300 lines of C.