News for hackers. Stuff that matters.
Re:Cursor Composer: Building a fast frontier model with RL by SafeDusk on Wednesday October 29, 2025 @08:15PM (#45754831)
I think both Cursor and Cognition and going in the same direction of SWE-grep[0].

SWE-grep was able to hit ~700tokens/s and Cursor ~300token/s, hard to compare the precision/recall and cost effectiveness though, considering SWE-grep also adopted a "hack" of running it on Cerebras.

I'm trying to kickstart a RL-based code search project called "op-grep" here[1], still pretty early, but looking for collaborators!

[0]: https://cognition.ai/blog/swe-grep [1]: https://github.com/aperoc/op-grep

Re:Cursor Composer: Building a fast frontier model with RL by swyx on Thursday October 30, 2025 @02:49AM (#45757079)
excited to see how far you get with opgrep!
Re:Cursor Composer: Building a fast frontier model with RL by SafeDusk on Thursday October 30, 2025 @06:15AM (#45758269)
hehe thanks! as a self taught AI engineer, might take awhile =D