Chris Wong

@codevsdev Books, I have a bunch with dust on them. If AI takes out the grid I can use them to start a fire in the winter to keep warm

how did people even learn to code when there was no docs, no YouTube... nothing?

@weswinder 🤣😂

fable 5 just refactored my entire codebase in one call 67 tool invocations. 1M+ new lines. 24 brand new files it modularized everything. broke up monoliths. cleaned up spaghetti none of it worked but boy was it beautiful

@jturntdev I found this out by mistake this past weekend. Funny thing was, while I was doing stuff, I thought to myself, man this is going really well and quicker...then i realized after an hour of two it was on medium. I just laughed. Clearly it's a thing.

I take it back. ChatGPT 5.5 medium reasoning with Extra high for planning. So much faster. Barely drains your weekly Quota. Can actually get in the flow of things. Feels like 2025 days. Thank me later.

@thsottiaux Like 10x my 20x haha Imaging the open source repos that could be developed with that.

I have a new kind of big button that I can press for Codex. Over the next 100 days, we will select one person per day who does impressive or incredibly useful work with Codex and give them 10X usage limits for a month to see what they can do with it. First one tomorrow.

@rileybrown Is it consistent usage rates cause that would be great...

Hmmmm any guesses?

@SullyOmarr soooo.. it's not just me? Besides usage, I've actually dropped reasoning to Medium and it's doing better 😂x.com/jorvekdev/stat…

no idea what they did with codex but gpt5.5 is awful for coding now literally unusable wtf

@thenowhereway Minimax...kimi...Grok...whoever.

If Cursor, Claude, and Codex all cost the same and provide exact same limits, which one are you opening first every morning?

@sflorimm Nintendo

Name a tech company that nobody hates

Is anyone else having this problems with Codex or am I building to much? I'm less than a whole day in and down to 59% left for week??? I don't use /fast and haven't done anything different than last week...

@enjojoyy Yeah, I’ve no idea. The longest I’ve gotten is just over 7 hours building an open source CRM repo (which still needs a lot of work haha) I used the plan to get a solid foundation then set goal to complete plan end to end. I can’t imagine what would even run for 24+ hours

People that run 24+ hours Codex tasks Can you share what you’re running exactly? Everyone is sharing the hours but not the task itself, I feel that most of them are just engagement baits

@LeaVerou @claudeai Just tell it, its work will be audited by Codex. Oddly enough the behavior changes.

💡Recent insight: gaslighting @claudeai seems to improve code quality >90% of the time. “You overengineered this, there is a simpler way” “There is a smaller delta that buys us most of the benefits” “There is a more elegant way” “This is not architecturally coherent” …before I even read its code. 😆

@thsottiaux Benchmarks are like the instagram of AI. The world is fed the cool, focused test, this is how awesome I am stuff. But then there's real life behind the scenes. I trust the real life stuff I hear from others where real problems are being solved in an effecient manner.

Do you still trust benchmarks or do you just listen to your friends? What makes you try a new model?

@sri9s Absolutely not

People using Codex, did you just switch back to Claude?

@thsottiaux @giordanorandone @OpenAI Anthropic would have totally shipped it, laughed, not answered customer service, then blamed Wile E. Coyote for dropping their Acme servers off a cliff trying to hit the Codex icon cloud

@giordanorandone @OpenAI Nah. The only thing that caught us off guard was a bug we discovered last minute. And we decided to not ship the said bug.

I am relatively sure that @OpenAI was caught off guard by Anthropic today and postponed the release unexpectedly. I cannot really explain it any other way, given how damn quiet it was, even though every Codex Thursday usually comes with announcements, even if they are only small ones. Today, there were a few small updates mentioned on the website, but they were not even properly announced on X. And when you also consider that there was vagueposting by OpenAI staff just yesterday, everything points in that direction for me. Are you guys disappointed, or do you think it is better this way?

@Thebiglade Time

What's more powerful than money ??

@thsottiaux You guys are just crushing it! Sometimes shit sucks when stuff like rate limits get messed up but this quick attention type service is why I switched over and keep it. Thanks!

Some of you noticed limits drained faster in Codex, we root caused it to an optimization that we rolled back that had an impact on cache hit rates when compacting across long running sessions. We fixed this and have now reset usage limits for all accounts. Enjoy the weekend.

@gdb I’m doing right now because of Codex draining my usage. It’s tedious, uneventful and probably pointless. I could probably just wait 3 days for weekly usage to reset and cover the three days in 3 hours like it never happened 😕

trying to remember what it was like to code before codex

@NoahKingJr Back to campfires and bourbon

What’s coming after Artificial intelligence?

Its like, do I drink the last cup of water and be thirsty for days, or ration it out and pretend I’m being hydrated. Or just go find a new well 🤷🏻‍♂️ Codex got me this week 🤦🏻

@rezoundous 😂 How do I test it when it burned up all my usage

Limit issue fixed? Anyone tried it yet?