I packaged up the "autoresearch" project into a new self-contained minimal repo if people would like to play over the weekend. It's basically nanochat LLM training core stripped down to a single-GPU, one file version of ~630 lines of code, then:
- the human iterates on the prompt (.md)
- the AI agent iterates on the training code (.py)
The goal is to engineer your agents to make the fastest research progress indefinitely and without any of your own involvement. In the image, every dot is a complete LLM training run that lasts exactly 5 minutes. The agent works in an autonomous loop on a git feature branch and accumulates git commits to the training script as it finds better settings (of lower validation loss by the end) of the neural network architecture, the optimizer, all the hyperparameters, etc. You can imagine comparing the research progress of different prompts, different agents, etc.
https://t.co/YCvOwwjOzF
Part code, part sci-fi, and a pinch of psychosis :)
Drag your movies from backlog to completed.
Do the same for books. And games. And TV shows. And podcasts.
BacklogBox — one Kanban tracker for all your media.
https://t.co/Nupwa0ChAh
@huntabyte Sorry for bothering, couldn't reproduce on new repo's and somehow it kept on falling back on using a non 2.0 Formsnap version even though it was declared as ^2.0.0.. Something messed up in my target repo. My bad, thanks for the upgrade!
@huntabyte Following the Quick Start on the new docs gives some weird behaviour. #snippet children({ props }) is still typed as any and the data gets erased/emptied as you post the form. Just straight up followed everything from this page: https://t.co/g5Ic9Is1pL
@dixpac@sylarruby@basecamp In case someone is looking for it, this article describes some topics. Other 37S blogs are good sources as well.
https://t.co/7Jyoi5I8GS