Rees
203 posts



@Leulaya @CastAsHuman honestly I don't think anyone does context compaction better than openai
English

@CastAsHuman because larger contexts eats up more compute which means reduced limits the compaction is truly a thing of beauty no need for 1M
English


Took us a minute, but stacked PRs are now on GitHub 🥞
github.blog/changelog/2026…
English

I have decided to move on from my past projects
I am ending my work on Lucia
I am deprecating Arctic
I am deprecating (most) NPM packages published under Oslo
pilcrowonpaper.com/blog/18
English

@YousufKalim @r0bert_rpg Is Cursor still doubling usage on 4.5 or has that ended?
English

@Jakubantalik love these!
there is a small bug where once the "Get Pro" screen shows for CSS/React then even non-pro transitions will show that same screen too until you refresh the page
English

Added 12 new transitions to transitions.dev 33 total
You can copy as CSS or React code or use as a skill for Claude, Cursor or Codex or use with Refine tool
npx skills add jakubantalik/transitions.dev
English

@bruvimtired is this the equivalent of it deleting the tests because they failed
English

When Claude writes code, it signs the commit: Co-Authored-By: Claude. I'm paying for the model, and it still takes a credit line. Okay, fine.
But co-authorship is a claim about contribution. And that logic should work both ways.
So what about all those people who involuntarily "contributed" the training data? You know, half the internet. Don't they deserve co-authorship on the models?

English





















