If the model "improves" every 5 hours, how do you have any guarantee of model co... | Hacker News

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

		htrp 24 days ago \| parent \| context \| favorite \| on: Improving Composer through real-time RL If the model "improves" every 5 hours, how do you have any guarantee of model consistency across long coding sessions?

annie511266728 24 days ago | [–]

Yeah, this feels tricky

If the model changes every few hours, we’re basically debugging against a moving target - and that gets expensive fast.

dbbk 24 days ago | | [–]

What do you need consistency for

Consider applying for YC's Summer 2026 batch! Applications are open till May 4
Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact