Sabitlenmiş Tweet
Tom Krcha
9K posts

Tom Krcha
@tomkrcha
Chief Vibe Officer at https://t.co/YJUz1TqYvM ✏️
San Francisco Katılım Ekim 2008
1.7K Takip Edilen21.2K Takipçiler

Try "Let it cook" feature in @pencildev, it creates multiple variations from the same prompt and iterates on it. What we discovered is that further iterations tend to get more creative/informed vs first shot. Also select a style as a starting point or drop in an image, honestly could be a photo, color palette or whatever, we call it "vibe image" ;)

English
Tom Krcha retweetledi

Is AI is making you a better author of your design work?
@tomkrcha paused with that question. He was giving it some serious thought.
He's been thinking about how to improve design work for decades:
- Was the youngest Adobe Flash evangelist
- Created Adobe XD
- Founded @pencildev
His thesis follows an emerging bet many design tool builders are making right now:
AI agents will increasingly share the canvas with humans.
We dig deep into what the future of design work might look like, and how he's building for it.
Don't miss this one.
English

“Sophisticated models” is the missing part really, the rest is already possible. But makes you wonder, why isn’t there a massive incentive to create a state of the art local model from the leading labs. Possibly because they don’t sell hardware.
But, Google could probably pull this off, they have a model, TPU and Google Home in millions of homes already.
English

imagine if apple basically let you set up a home “server” that ran inference on that device with sophisticated models & every apple ecosystem device is a node off of that central server.
it’s complicated but if anyone can deploy hardware like this it would be them. this would allow zero marginal cost ai without any middle man with a super privacy first approach.
it would create another product category entirely. anyone remember the airport extreme?

English

Real-time World Models are the next AI frontier.
Today, we're taking the first step towards this reality: our early preview lets you experience worlds generated in real-time, running on our global low-latency infrastructure.
Try it now: reactor.inc
English

Today, @editframe emerges from stealth. Agents need video.
Editframe Agent Skills:
npm create @editframe@latest
Just prompt Claude Code, Cursor, or Codex and get a working video or a full interactive GUI.
This video was created just by prompting 👇
English

@Ignaci0m_ you can use Illustrator or Figma, both work great, there are also some online tools
English

Imagine every pixel on your screen, streamed live directly from a model. No HTML, no layout engine, no code. Just exactly what you want to see.
@eddiejiao_obj, @drewocarr and I built a prototype to see how this could actually work, and set out to make it real. We're calling it Flipbook. (1/5)
English

@argostore_sk @petergyang why do all these online tools look like they are gonna rob you at daylight
English

@petergyang @tomkrcha This free vectorization tool works best with black-and-white images: vectorization.eu However, you need to upscale your bitmap image and convert it to JPG first. There are tools for that available on the website.
English

@rileybrown It totally has a transparent background, in fact, it's perhaps the only image model that has it. Also, another trick is to vectorize it x.com/tomkrcha/statu…
Tom Krcha@tomkrcha
One of the interesting use cases with ChatGPT Images 2 is that it can generate high-resolution logos that can be easily vectorized/traced into SVGs with reasonable quality.
English

@petergyang @brycedriesenga @QuiverAI There are plenty options for the vectorization, all different qualities, resolutions, etc.
English

@petergyang there are models, but you can also use trace image in Adobe Illustrator or vectorize in Figma.
English

@tomkrcha I'd love to see a screenshot with vertices displayed.
English

Pica is a fully native app for managing your fonts on MacOS.
Organize into collections, test color themes and logos, watch folders, manage what's installed, and much more.
Available for free at pica.joshpuckett.me
English

I feel like Codex's gap in frontend design skills can be easily made up if you use an AI design tool.
My favorite is @tomkrcha's Pencil
English







