# autophys

- **Event:** [Meta Global AI Developer Hackathon](https://cerebralvalley.ai/e/meta-ai-developer-hackathon)
- **When:** Oct 26, 12:00 PM – Nov 4, 9:00 PM (EST)
- **Where:** Online
- **Team:** [James Kao](https://cerebralvalley.ai/u/yuzuquat)
- **GitHub:** https://github.com/yuzuquats/autophys
- **Demo video:** https://youtube.com/video/PfxEHRruaLg
- **Gallery:** https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4733

Got a physical model and some messy measurements? You can fit the model until the curve looks right, but that doesn't mean the numbers mean anything. Half the time, a totally different set of parameters fits just as well, and you'd never know from looking at it.

AutoPhys checks that for you. It fits your model, then tells you which parameters your data actually nails down — and which ones are still basically guesses. If you're missing something, it'll suggest what to measure next to sort it out fastest.

You run the test (or let an agent propose one, and sign off on it yourself). New data goes in, it refits, and checks itself against data it held aside. Everything — every fit, suggestion, decision, and measurement — gets saved locally so you can go back and see exactly what happened.

Under the hood it's a Python SDK for simulation, fitting, and figuring out what's identifiable, with battery, thermal, and oscillator models included. There's a Rust daemon and a browser workbench if you want to run it as an ongoing investigation, plus SDKs, tests, and examples to poke at.

It won't make up physics for you. It's just there to help you get a straight answer out of the model you've already got, without running a hundred experiments you didn't need.

i worked on this project with jnlucero96@gmail.com but wasn't able to add him to the roster.

## More from Meta Global AI Developer Hackathon

- [formicslab](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4730)
- [NOVUS](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4731)
- [Gaurav Bharatavalli Rangaswamy](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4732)
- [Critics wonder](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4734)
- [SkyNet](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4735)
- [George Sung](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4736)

---

Markdown version of https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4733. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt