The CEO Vercel is making web components. No framework dependency, using the web platform as intended.
I’m beginning to believe
Got to experience the magic of @uisdk.
Two prompts of fx + Gemini Argon. Pure Web Components. No framework.
The agent reused 14 UI SDK behaviors out of the box 🤯: window resizing, drag, menubar hover-switching, typeahead, roving toolbar focus, 10-band EQ sliders, scoped hotkeys, ARIA, and WebMCP agent control.
winamp-ui-sdk.vercel.app/ (source linked)
Music by @suno.
19
JB Cornac retweeted
the most import skill in 2026 is still being able to tell your agents "wtf is this crap, no, not like this"
if you stop doing that, you're basically fucked, despite what some devfluencers would like you to think.
JB Cornac retweeted
New rule: never use React SSR.
I convinced myself that things would be different this time, but SSR still sucks. It adds a ton of complexity, increases your bill, and solves no problems (for apps).
DHH wrote "When wolves get out of control, you shoot them. When gypsies take over public spaces, you deport them." and these people shrug.
But a Linux project switches to the GPL, the same license as the kernel, and they all lose their minds.
Real people vs. a license file.
Replying to @asmirkn
That sure aint the reason they moved to GPL days after Omarchy integrated and credited a bunch of Asahi’s work.
holy shit
We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math.
It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training.
We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵
Blog: rikiyat.github.io/blog/weigh…
Readers added context they thought people might want to know
LLM was given weights explicitly as text in prompt and used code/forward passes (per authors' logs) to ablate Qwen looping direction, a known thinking-mode issue per model card. Primarily fixed repetition in that mode, not general 'natively' improving weights/math.
rikiyat.github.io/blog/weight-sp…
github.com/RikiyaT/lm_und…
huggingface.co/Qwen/Qwen3.5-0…
huggingface.co/RikiyaT/Qwen3.…
82
JB Cornac retweeted
Replying to @unclebobmartin
“Your job is to help the LLM write better code faster.”
Strange: I thought the job of a software engineer was to bring value to stakeholders in an enduring, economically sustainable, and ethical fashion.
I suggest you reconsider your life choices, Bob.
JB Cornac retweeted
Im sorry im not buying this ‘not reviewing code’ thing. Maybe later, but certainly not now, even with frontier models with max thinking
Every single day I’m catching nonsense. If you’re merging this shit without looking, the systems your working on are either
- not important and/or
- have zero consequences for being wrong
JB Cornac retweeted
"Why is formal verification exploding NOW?"
1. Nobody understands the code LLMs write
2. Everybody hopes for some deus ex machina
JB Cornac retweeted
> it is very very likely that fable is better than you in every branch of software
It is faster than me. It is cheaper than me. It knows way more trick than I do. It also indeed knows every codebase that is out there.
But for some reason, it's dogshit at actual engineering. Tries to cheat its way through problems instead of fixing them constantly. Completely fails to produce sensible abstractions. The test suite it produces are both enormous and somehow misses the behavior you care about while being tightly coupled to the implementation.
I wish it could do these things for me, but it can't today, and doesn't show much signs of improving.
I think people saying this are either doing it for the views, or because they are the ones who produced the mountain of shit software we ase dealing with are are indeed worse than fable in every way.
Not something I'd brag about.
the coding agent haters are getting dunning-kruger'd so hard right now
do you really think that your engineering ability is better than the models at this point?
it is very very likely that fable is better than you in every branch of software
it has been trained on all the code of the internet, it knows almost every trick in the book, all the design patterns of every domain, every single low-level detail, every bit-twiddling hack, it's read every single blog post from every cracked engineer that ever existed
do you really think you can beat it? do you really think you can be more creative than it?
unless you are 99th percentile expert in some niche domain of software, it simply is not possible. sorry.
JB Cornac retweeted
You guys disgust me, honestly. If you think you can't compete with today's actually retarded AI models you have not done anything hard enough. Fortunately the AI can teach you how to surpass it if you just get off your pathetic whiny asses and put in a little effort
AI can rapidly generate trivial filler code, but if you go completely hands off it very quickly goes into a doom spiral and makes completely useless trash.
Just check out the photoshop killer ArtCraft, the web browser from first principles someone made, literally every vibe coded web game, etc. Or AI art for that matter, it generates boring garbage unless you take the reins and steer it with your detailed artistic vision. AI is great at what it can do but it's a tool and its capabilities are limited.
Keep your hands on the wheel and set your sights on something more ambitious than the pathetic little web dashboards and scripts and applets you've been one-shotting
the coding agent haters are getting dunning-kruger'd so hard right now
do you really think that your engineering ability is better than the models at this point?
it is very very likely that fable is better than you in every branch of software
it has been trained on all the code of the internet, it knows almost every trick in the book, all the design patterns of every domain, every single low-level detail, every bit-twiddling hack, it's read every single blog post from every cracked engineer that ever existed
do you really think you can beat it? do you really think you can be more creative than it?
unless you are 99th percentile expert in some niche domain of software, it simply is not possible. sorry.
JB Cornac retweeted
if i take a walk in my little walkable neighborhood to buy fresh flowers and crunchy nectarines and i see your tesla loudly flip in the air and destroy my serenity, i will blow up your car with a rocket launcher
JB Cornac retweeted
I feel envious of all you people who work on simple and boring enough problems that you think AI is better than you at architecture
I WISH it was better than me at making architecture decisions man but it’s just plain dumb at that stuff
it’s clear we need to retire the term vibe coding because the models now produce better architecture & way more coherent code than an average software engineer.
perhaps even better than top software engineers too because unlike a human, ai has a much much larger context window.
this ain’t vibe coding at all. it’s precision engineering.
JB Cornac retweeted
I’m still calling bullshit on the software factory bois.
I’ve not written code since March. I’m using Codex, Claude, and Grok Bot non stop through the day; at home, in the cab, at the gym. I’m building stuff people actually use. Not millions but tens of thousands.
And I’m still not running 20 agents in parallel for 8 hours at a time.
And every time some dude tries to prove me wrong , he shows me a piece of shit generic software factory or social media scheduler. It has no users and the sloppiest of landing pages (“coming soon”).
It still takes a LOT of work to drive what to say no to and to sweat the details.
Talking to smarter folks than me, I'm convinced many of the AI folks in my timeline are full of shit.
Nobody is "running 20 agents over night" and building stuff for actual users. Maybe some are building internal tools or disposable software. Maybe.
But building software people like using? That doesn't get hacked on day one or blow up after the 3rd user? Nope.
I don't even understand what that's supposed to look like. Do you work out a 57 pages document that perfectly describes what you want to build and then summon 14 agents and have them run wild for 6 hours? And what comes out on the other end isn't a broken pile of shit?
Nope. Not buying it.
PS: it may also be that I have an IQ of 82 and can't figure it out.
JB Cornac retweeted
Replying to @ImigrantPatriot @k_kohlbrenner
being an early adopter doesn't turn this into harry potter
JB Cornac retweeted
They certainly don’t produce better architecture or more coherent code that the average engineer I’ve interacted with
I’ll stick with calling it vibe coding ty
it’s clear we need to retire the term vibe coding because the models now produce better architecture & way more coherent code than an average software engineer.
perhaps even better than top software engineers too because unlike a human, ai has a much much larger context window.
this ain’t vibe coding at all. it’s precision engineering.
JB Cornac retweeted
We’re witnessing the largest regression in software engineering maturity ever seen.
JB Cornac retweeted
This is like putting weak brakes on a sports car.
Unit and integration tests report regressions *quickly*.
E2E tests are too slow to cover all scenarios. I’ve seen teams try E2E only. Their tests ran for over an hour.
To move fast safely, you need to be able to stop quickly.
JB Cornac retweeted
Yeah he’s going on a celebrity tour about how he one shotted a port like many of us did in 2025 but we threw it out because it was bad.
He doesn’t even know it’s bad.
Replying to @kellabyte
When someone suggested that he’s doing something wrong with Go as the numbers don’t make sense, his response was “It’s open source, you are welcome to a PR”.
Does anyone really have time to dig into AI slop to figure this out?


























