Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Is this like Weights & Biases for prompt-based LLM experimentation (as opposed to model training)?


That's not a bad description. I don't have a good comparison— we mostly just built this as a tool that we personally needed. I like to think of it like W&B but also in the world of logging tools like Prometheus.

LLMs are a funky thing to work with, and we need a whole suite of new tools to make them prod-ready.


Sounds like it, pretty smart idea




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: