Cheng Lou ✓
@_chenglou · California, USA · joined 11 Jan 2015
Worked on: @reactjs, @messenger, @reasonml, @rescriptlang Currently: @midjourney, Pretext, Freerange
71 054Followers
501Following
8 703Posts total
30.3MViews on collected posts
Latest posts
Looking for new posts…
More poems at https://t.co/uQJJ8hXRzH
New AGI benchmark?
2.6K views · 24 likes
· 0 reposts · 1 replies
Open on X →
And finally, this one is readable in all → ↓ ← ↑ directions: https://t.co/aQfpEYYT3V
3.4K views · 36 likes
· 3 reposts · 2 replies
Open on X →
This one produces the same (coherent) poem horizontally and vertically: https://t.co/6K4TUp75zR
2.8K views · 34 likes
· 0 reposts · 1 replies
Open on X →
I’m proud to announce Hengshu, where instead of testing GPT Astra properly, I’ve whipped it to create omnidirectional Chinese poems 📜
Here are 3. First one rhymes and reads both horizontally and vertically: https://t.co/i6KEUNvDLP
44.9K views · 776 likes
· 66 reposts · 40 replies
Open on X →
Hey folks. Sotaku 2 is out! Timely, given the speculation about GPT Astra’s recurrence
Tldr: Sotaku is a tiny laptop-grade neural net that learns to solve sudoku without being given its rules. The broad goal is to see if we can solve bigger symbolic problems purely neurally, and
17.8K views · 277 likes
· 23 reposts · 13 replies
Open on X →
50% chance this guy’s a generational poster
61.3K views · 1.5K likes
· 5 reposts · 7 replies
Open on X →
Petition to replace all white rice with biryani
50.1K views · 1.1K likes
· 56 reposts · 100 replies
Open on X →
Remember how back in the crypto days folks asked "why can't we make these GPUs run useful work" and the typical answer is "making GPUs run useful work for money would defeat the purpose" and it turns out that's completely wrong in an important way
5.9K views · 72 likes
· 4 reposts · 3 replies
Open on X →
https://t.co/2meoiB2nFj
Run it through your codebases and give some feedback!
8.4K views · 212 likes
· 8 reposts · 9 replies
Open on X →
I invite you to re-examine your own domain carefully and see whether a tipping point is already here for you: how much of what we do, of our existing “best practice” patterns, should be rewritten to leverage a sufficiently smart coder? Ask your agent; you might be surprised!
7.7K views · 37 likes
· 2 reposts · 1 replies
Open on X →
As the quantity of code explodes in the next few years, I hope the folks worrying about quality can rest assured that there’s evidence that our ability to verify even the most casual vibe code can keep up with the amount of slop
Turns out that math was the ultimate spec
7.7K views · 47 likes
· 2 reposts · 1 replies
Open on X →
As a second order effect, the more we use tools like Freerange, the better the AI models become, because we’d feed it more correct code data & better reinforcement learning. What better to teach to AI, if not more practical maths?
6.6K views · 41 likes
· 0 reposts · 1 replies
Open on X →
The first day we introduced Freerange into our Midjourney codebase, we’ve caught 12 layout bugs that slipped through the best agents. Can’t really blame LLMs for not catching those across thousands of files and shaky floating point calculations.
7K views · 48 likes
· 1 reposts · 1 replies
Open on X →
But AI doesn’t care; it’ll cut through the thick layers made for humans and rewrite code as fast and as well as needed
The sweet spot of what constitutes “good code” is gradually shifting in the age of AI; we’ve been having delayed reactions
7.6K views · 38 likes
· 1 reposts · 1 replies
Open on X →
The idea of Freerange isn’t new at all; powerful provers exist already in the wild. Google’s AlphaProof gets smarter the more verification they threw at it. But pre-AI, it was rather unrealistic to ask a dev to:
- statically verify some dirty scripts when quickly iterating on
7.6K views · 53 likes
· 1 reposts · 1 replies
Open on X →
Another one, where we found a clash between your navigation bar size and the mobile site’s width. No more ugly mobile websites with bad column sizings! “number -> number” is the new “string -> string” https://t.co/pTkSrgVQKY
8.5K views · 54 likes
· 0 reposts · 1 replies
Open on X →
Consider this progress bar logic. What do we know about start and end here? Well they shouldn’t be equal. Otherwise you’re dividing by zero and BAM, you’d be carrying around an Infinity that pollutes everything downstream. Freerange audit catches this https://t.co/VRQGDRR90O
10.2K views · 55 likes
· 0 reposts · 2 replies
Open on X →
Your app has lots of numbers. If you ask “what do we know about these numbers”, most programming languages will tell you “not sure” or at best, “these are integers”. This doesn’t help you nor the AI agents
10.9K views · 64 likes
· 1 reposts · 1 replies
Open on X →
My dear UI developers, ML practitioners, and fans of programming language:
Many months & billions of tokens later, I’m proud to present to you the first step in our long, long collective journey to turn vibe coding onto proof engineering, starting with: making user interfaces ht
192.7K views · 1.5K likes
· 95 reposts · 42 replies
Open on X →
🚨 Hello! This post reached beyond its original audience. If you're wondering why you'd want dancing balls while reading: you don't. It's a demo to showcase the expressivity & performance of the system for designers & engineers
For immediate benefits, see https://t.co/vN1kycwfst
118.7K views · 684 likes
· 29 reposts · 24 replies
Open on X →
Latex fans assemble! It's time to use Pretext's expressive controls to improve text readability.
@Somnai_dreams implemented the Knuth-Plass algorithm to reduce reading churn on long paragraphs of text: https://t.co/XjPyegmGnu https://t.co/owdOyWO2AK
355.8K views · 938 likes
· 81 reposts · 28 replies
Open on X →
0:09
@_chenglou It's time to bring these to the web https://t.co/qJdVsCNOT3
111.8K views · 2K likes
· 33 reposts · 12 replies
Open on X →
Fun glossary of demos. Some useful, some purely showing off:
https://t.co/wFzfFirpnq
https://t.co/01suDjnPpd
Will keep adding more!
277.9K views · 1.3K likes
· 55 reposts · 45 replies
Open on X →
Project’s at https://t.co/meTmVhJXlk
`npm`/`bun install @chenglou/pretext` and throw your AI into making cool demos for it!
351.7K views · 4K likes
· 191 reposts · 49 replies
Open on X →
This paradigm is significantly more performant (~500x, but it's an unfair comparison; read on) than laying out text and messily crawl out the measurements through getBoundingClientRect(), which causes DOM read/write interleaving, one of the most expensive operations on the web
300.4K views · 885 likes
· 16 reposts · 1 replies
Open on X →
The engine’s tiny (few kbs), aware of browser quirks, supports all the languages you’ll need, including Korean mixed with RTL Arabic and platform-specific emojis
This was achieved through showing Claude Code and Codex the browsers ground truth, and have them measure & iterate
309.1K views · 1.4K likes
· 33 reposts · 15 replies
Open on X →
5. Your typical auto-growing text area, accordion, multi-line text centering, pure canvas multiline text, and all other things that used be real CSS challenges, now reduced to a boring footnote https://t.co/j3JRgcP37a https://t.co/1teTkfm5th
324.1K views · 1.3K likes
· 23 reposts · 8 replies
Open on X →
0:10
4. Variable font width ASCII art, because why not, it's easy now https://t.co/bDyARIwdqF https://t.co/BgzGYmOjTC
351.6K views · 2.6K likes
· 70 reposts · 8 replies
Open on X →
0:12
0:09
Against accounts of the same size
9 posts from the last 90 days, next to the 10K–100K follower range. shown to more people than peers of the same size.
Median views5 879this account924median for 10K–100K
Reach, %8.27%this account3.62%median for 10K–100K
Engagement, %1.38%this account1.52%median for 10K–100K
| Metric | This account | Median for 10K–100K | Ratio |
|---|---|---|---|
| Median views per post | 5 879 | 924 | 6.36× |
| Reach (views ÷ followers) | 8.27% | 3.62% | 2.29× |
| Engagement rate | 1.38% | 1.52% | 0.91× |
Others in this range → Compare with another account → How these benchmarks are built →
Growth & engagement
How the posts we collected actually performed: views and reaction rate post by post, what the audience did with them, and where the follower count goes.
Views per post
6.6K21 Jul
7.7K
7.7K
8.4K
3.8K
5.9K27 Aug
50.1K1 Sep
61.3K
17.8K2 Sep
44.9K4 Sep
2.8K
3.4K
2.6K
1.1K
Last 14 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.
Engagement rate per post
0.63%21 Jul
0.66%
0.52%
2.71%
0.19%
1.38%27 Aug
2.51%1 Sep
2.49%
1.76%2 Sep
1.99%4 Sep
1.26%
1.22%
0.98%
0.97%
Reactions — likes, reposts, replies and quotes — divided by views. Median for 10K–100K accounts is 1.52%.
What the audience does
Likes64.4%21 112 in total
Reposts2.4%799 in total
Replies1.3%421 in total
Quotes0.4%147 in total
Bookmarks31.5%10 314 in total
Share of every reaction we collected for this account. Replies mean argument, reposts mean endorsement, bookmarks mean the post was worth keeping.
The follower curve appears once this account has two daily snapshots — we take one a day, and this one is on its first.
Similar accounts
Barun Sobti
@BarunSobtiSays
142.1K followers
· 127 posts
Incognito ✓
@Incognito_qfs
142.1K followers
· 77.3K posts
Where Winds Meet ✓
@WhereWindsMeet_
142.1K followers
· 1.3K posts
Dr. Osman Aşkın Bak ✓
@OA_BAK
142.1K followers
· 20.3K posts
Graham Rahal ✓
@GrahamRahal
142.1K followers
· 33.2K posts
Choptimus Prime
@mikesmicYT
142.1K followers
· 6.7K posts
Xico Graziano ✓
@xicograziano
142.1K followers
· 48K posts
PA Media
@PA
142.1K followers
· 160.3K posts