Walid

if i was @DarioAmodei i will offer @thsottiaux 1 billions a year before it's too late

@thsottiaux @robinebers I can't understand you have coedx with unlimited token you should ship more than that probably *1000 more

@robinebers Excuse our shipping

Codex now has a Security Scanner? when tf was this added and why is this not a bigger deal?

@wan73184085 @thsottiaux @jxnlco high without fast.

@thsottiaux @jxnlco I ran out of week usage just 3 days, pro account

/goal might be the most consequential thing we have shipped in codex The value of good instructions has never been higher.

@corbin_braun skill issue

GPT-5.5 is amazing. but please don't use it for the frontend.

@melvynx you switched to codex now

agent is Codex app is beautiful 🤔🤔

@sama @official_taches 40 usage 300 a month please 😁

@official_taches what do you need for just one plan?

I’ve officially cancelled both Claude Max plans and have 2 x Codex Max plans. Codex - particular GPT5.5 is the best coding model.

@damnGruz codex app is pretty cool, but the fact that i use both claude and codex i feel the terminal is more adapted for me

has everyone switched to GUIs?

@sama @sama Can we get something like high reasoning for planning and medium reasoning for implementation, similar to what Anthropic did with Opus and Sonnet?

5.5 xhigh in fast mode is really good i think i got psyoped by twitter on medium for a bit

@sama Yes, but the limits are much tighter now, so nobody is using it anymore.

@mttcer The world has changed, my friend. You don’t need to know every single line of code anymore. But you do need to automate tests so you can be sure old features aren’t broken. Either you adapt and adopt this, or you’ll be replaced. Hard truth, sorry.

not a critic to Max, but i still don’t get how people can let agents run this long. for me, i use agents only for planning / coding individual features, but i still need control over the overall codebase and decisions i feel like that if i let an agent run for hours i’ll come back with a messy codebase that would take days to understand and review

@uccmen no only Anthropic for now.

@walidbelr wow TIL. is there something similar for OpenAI/Codex?

Become a Claude Code Architect from Anthropic: anthropic.skilljar.com/claude-certifi…

@embirico @thsottiaux @ajambrosino @thsottiaux please kick him out him from the team

okay what a week. @thsottiaux @ajambrosino let’s avoid shipping anything significant until next week?

@jasperdevs @thsottiaux I think he represent openai now more the sam altman. The true ceo.

this dude has single-handedly brought OpenAI up about $20 billion since he started tweeting @thsottiaux has too much aura at this point

@ajambrosino Working in two sessions in parallels ?

Codex for everything: - Dynamic UI for the task at hand - 20% faster computer & browser use - Even better slides and sheets - Annotate in browser, artifacts, and code - Easier to get started - Cleaner design across the app - Performance improvements - (no clunky handoff/switching)

@b_nnett it didn't work for me in the latest version how to enable it ?

@walidbelr already a tweak!

iOS simulator directly in Codex!

@robinebers @noisemakerjon superior to gpt 5.5?

@noisemakerjon been saying this a lot Opus 4.7 is the superior model for most people right now. just not in Claude Code.

Opus 4.7 fixed something for me that the last 4 checkpoints of GPT couldn’t

@CastAsHuman @dkundel vpn

@dkundel From the top of my head, the only thing missing right now is computer use in Europe. Don't wait to try it!

I love Codex so, so much!

@sama you love the money of the users but we love the quality of your models

we love our users

@fcoury @fcoury computer use in the cli

Let me take this opportunity and ask CLI users: what are your small papercuts?