Tedi

@yacineMTB Codex is definitely the primary driver, but I'm still keeping claude for design/front-end taste updates.

There isn't a single smart person I know that uses claude code over chatGPT 5.4 xhigh by the way. The only reason anyone would use claude is, amusingly, because claude does not have guard rails. But now they do, so there is no reason remaining to use it

@O_Pawica This looks great Oskar. Are the elements in the header tappable/can you scroll by dragging the items in the header? I've got a similar solution but those conditions are wonky.

React Native Header Motion v1 is out 🎉 Build scroll-driven animated headers your own way! 🖐️ New in v1: header panning Users can drag the header itself to scroll, making the gesture feel continuous with the scrollable. 🗒️ pawicao.github.io/react-native-h… More details in the thread 👇

your db query is slow? just add this! boom, now you are ai-native instead of lazy!

@chamath If you have your AI chats on your machine via something like codex, claude code, etc, you can easily create a watcher for each of their chat history folders and have it extract int your knowledge base. You can then run tasks periodically to summarize/cache based on your needs.

This may be a dumb question but I’ll ask it here anyways: I can’t find a good way for my various AI chats to automatically sync its conversation history into a structured knowledge base. So that as I update various chats from time to time and refine context, my knowledge base automatically grows with this new info.

Clear my evening

@ElishaDLong Retardmaxxing is the 2026 version of the ubermensch

@InsomniacPM22 💯

quality of context > quantity of context, every time. people assume "more tokens = more info for the model" but the model still has to figure out what matters inside that window. the harder you make that job, the worse the outputs get. relevance filtering before you even hit the API is the real unlock

Just because there's a 1M context window, doesn't mean it's ideal. Regardless of window size, performance degrades quickly as you climb up the ctx size.

5.2 extra high is the wise oldest brother. Slow but careful and smart. 5.4 is the young ADHD brother. Super smart and eager, jumps all over the place, and talks 100mph. Lots of potential but needs babysitting from the older brother. 5.3 codex is the middle hard working sibling.

I don't trust benchmarks, but it definitely feels like this. Had to stop using it the past week for anything other than easy interaction problems.

@thsottiaux Ability to group conversations into folders!

Speed and front-end capabilities: we heard that loudly, and things are improving quickly, to the point where I think we are now highly competitive on speed, if not best-in-class. What else should we improve for Codex?

Hey @grok, block all tweets that have "it's so over in them"

@jess_ann_pin 40s feels great if you take care of yourself.

Men drastically underestimate how much worse looking they get after 40.

Hey fellow nerds, it's time to start livin'

Memory is not always useful. Clean slate = better objective feedback. I want my friends to remember but not necessarily my coding agents.

@mehulmpt most developers are not “good” developers

the more I use opus 4.5 to build personal apps locally, the stronger I feel this will NEVER replace good developers. the code and practices it creates on its own, without supervision, is basically garbage that works.

@rawmilkhoney hits all the food groups

Sometimes dinner is just half a pb&j and some sausage on a paper plate

It's so funny to me that people are off of work already, and think they're going to lap their competition.

still hitting records at 41, let’s see how long we can keep this going