Joined January 2026
- Tweets 1,195
- Following 283
- Followers 547
- Likes 2,631
152 Photos and videos
Pinned Tweet
I got to edit the latest Dylan Patel intro!! What do you think?
5
8
985
Honestly I feel limitless now. Matt's cooking with this video - shows how easy it is
Jun 12
I built a custom Bitzee/Tamagatchi with my son.
It was SUPER easy. Vibe coding feels like cheating.
2
73
"Fluff" in nonfiction is a matter of disposition.
If you get mad every time you read something you consider fluff, then yeah, it'll be worthless to you.
But fluff is AWESOME. Extra stories and examples provide memory hooks for your brain to recall the lesson. More time spent reading lets your brain l sponge up the book more effectively.
Don't fall for downloading book summary apps that promise value
1
22
Nintendo was cooking when they wrote the anthem of fat people
(Wii Fit - Obese)
1
50
Codex pet feature has potential, but its current implementation is needlessly restrictive. There's no ability to control frame timing. There's no ability to add animation logic.
On top of that Image 2 is frustratingly unable to generate true-pixel images.
It's actually so close to great... but it needs more.
1
1
71
I was having a good time creating animated states for houndoom before I realized that Codex Pets wouldn't be able to handle it. Here he is falling asleep.
33
I've been doing this. Feels horrible. Like I can see my brain becmoing stupid in real time
Jun 10
only boomers fix typos in prompts. llms perfectly understand you even if you mistype.
3
130
quote tweet
62
AI has NEVER one shot anything useful. Stop testing based on one shots
14
2
28
2,423
When does AI start making useful stuff instead of just burning electricity to redo work that's already been done?
Jun 10
So I gave Fable 5 the watchmaker benchmark: a full Swiss lever movement in Three.js. Real gear ratios (18,000 bph), working escapement, breathing hairspring — and the hands tell actual time. It verified its own work with vision, in a loop, until done. ⌚
2
5
762
I will now be PERMA BLOCKING anyone who uses AI to write
nothing worse than getting a few sentences deep into something that not even the author cared enough to write themselves
1
1
96
Microstrategy (MSTR) had to fall for Bitcoin to work. AT ALL.
How stupid is it to try to hoard ALL of a currency while still pushing the narrative of decentralization?
If it worked, Saylor would have been a veritable god
50
I've superpowered my life with AI
After months of practice, I'm proud to say that my personal skill has atrophied to the point where today:
I can write 4 minutes worth of code in 12 hours.
AGI AGI AGI don't get left behind
2
157
This guy literally built a website that predicts if Codex will reset limits
Jun 4
Will OpenAI reset Codex quota soon?
Inspired by @theo's prediction this week, I built a Super Scientific Methodologyâ„¢ to predict when @thsottiaux will press the quota reset button next.
Prediction signals:
🔹OpenAI team vague-posting
🔹API/Codex downtime/issues
🔹Potential feature or product release soon
🔹OpenAI parties or events soon
willcodexquotareset.com
2
23
9,051
They (AI agents) may imitate language, behavior and analytical skills, or even simulate empathy and understanding, but they do not understand what they produce, for they lack the affective, relational and spiritual persective through which human beings grow in wisdom.
Pope absolutely spitting 🔥
1
1
83
🤣🤣
2
82
NEVER, never leave your agents unsupervised.
This is crazy (GPT 5.5)
23
can some of you please uninstall codex? it's starting to crawl
1
53