Zac Morain
9.5K posts
@makeitmorain
just another nerd ruining baseball with some football on the side; Formerly of @DrivelineYouth & @Drivelinebb; @802_lab co-founder & former Director of Pitching
So the other day I had an idea.. and with a sick baby giving more working hours, I gave it a shot. Anyways I saw this the other day and was intrigued. I had a couple of ideas, but not the power to run much locally. x.com/karpathy/statu…
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. github.com/karpathy/autor… Part code, part sci-fi, and a pinch of psychosis :)
I spent parts of the last few weeks vibe coding baseball tools (what a weird sentence to write). Shocked at how much I'm now actually using what I built. Made a video walking through the tools (which YOU can use!). More details in thread, tease below!
That, either way, time would pass. Things change. There are things to learn. I thought I was going to fall further behind, leaving me without the skills to come back to a baseball role somewhere. So why does that change anything? Time will passes, so do something with it.
I drove 3,300 miles across the country, to a town I didnt know anything about with my girlfriend (now wife) all because I knew that was the place to be if I wanted to put myself in the best opportunity to succeed As @drivelinekyle says… “the time will pass anyways”
Indeed. Cignetti’s recent interviews stand out on not overworking yourself but it’s typically required from a young age (in your 20s-30s) so you can accelerate your growth into mid-career faster than your peers. Remember - the time will pass anyway. Use it. It waits for no one.
It cannot be overstated how seamless the operation of our launchpad is. When explaining to external staff, we always get the “that’s it?” when we explain the operation instructions. Yes, it is that simple. Designed so that ANYONE can operate it. Let our pipelines do the heavy lifting. Easy and clean interface and with one button click an assessment is completed.
