Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

The next prompt: Now convert it to javascript and WebGL. Make it responsive and performant, so I can resize the window. The room needs to adapt in real time. Still no libraries.

fofr

91,267 subscribers

36,109 views • 20 days ago •via X (Twitter)

Science & Technology

Anya Rossi• Live Now

Private livecam show

10 Comments

fofr20 days ago

Code:

TomLikesRobots🤖20 days ago

Oh wow. That kind of responsive design is what I used to dream of when I was a web dev.

fofr20 days ago

Yes! I was nostalgic for the early days of responsive web design too. I was thinking, why shouldn't a room reflow and adapt like a website does?

genevieveh@20 days ago

this is insanely impressive

Stern - uɹǝʇS20 days ago

I like this

Google20 days ago

Stretching the limits 4 real

Kyrelle Buckley20 days ago

is it still ray tracing in the shader or did it switch to raster

Julian C20 days ago

Bet the window resizing will get spicy.

Guillermo20 days ago

still no libraries and it has to resize live. did the lighting survive the port from python or get simplified

@Jeuri20 days ago

Prompt tip: when you chain "no libraries" + WebGL, lock one constraint per step (geometry first, then resize, then perf). Agents thrash when you change three rules at once. Did the room adapt without a frame drop?

Related Videos

“Since the time of Fela, Bob Marley, No one has ever cut across the way they did it. So my next plan now is to take it global. Someone needs to do and I feel I can do it.” -Wizkid (few years later, he did it)

“Since the time of Fela, Bob Marley, No one has ever cut across the way they did it. So my next plan now is to take it global. Someone needs to do and I feel I can do it.” -Wizkid (few years later, he did it)

Arthur🇳🇬🇬🇧🇸🇳

91,322 views • 9 months ago

Watch this robot dog learn to walk from scratch in real time! Our new method, APRL, dynamically adjusts exploration constraints to enable fast and performant RL directly in the real world. APRL can also adapt to changes in the terrain. No simulation, no demos. A thread 👇

Watch this robot dog learn to walk from scratch in real time! Our new method, APRL, dynamically adjusts exploration constraints to enable fast and performant RL directly in the real world. APRL can also adapt to changes in the terrain. No simulation, no demos. A thread 👇

Sergey Levine

105,579 views • 3 years ago

Did you know that you can create recessed screw holes right in PrusaSlicer? Add a negative cylinder, resize it, copy and paste it again so it will be aligned and placed exactly in the same spot. Resize the cylinder to make it bigger, position it, and you’re done! 🤓 More info can be found here:

Did you know that you can create recessed screw holes right in PrusaSlicer? Add a negative cylinder, resize it, copy and paste it again so it will be aligned and placed exactly in the same spot. Resize the cylinder to make it bigger, position it, and you’re done! 🤓 More info can be found here:

PrusaSlicer

13,926 views • 2 years ago

Today I completed this Fully Responsive X Clone Made this with just HTML and CSS. It took me 3 days to complete it and I am very happy with the results. 😁 I think its time to move on to bigger challenges now! JavaScript here I come 🚀

Today I completed this Fully Responsive X Clone Made this with just HTML and CSS. It took me 3 days to complete it and I am very happy with the results. 😁 I think its time to move on to bigger challenges now! JavaScript here I come 🚀

Shubh Sharma

87,258 views • 3 years ago

VFX-JS supports HTML-in-Canvas now. You can turn HTML element into a WebGL texture. It automatically detects the browser and fallback to the old SVG foreignObject workaround. So it works on Chrome, Firefox, and Safari, of course! #WebGL #VFXJS #Codepen

VFX-JS supports HTML-in-Canvas now. You can turn HTML element into a WebGL texture. It automatically detects the browser and fallback to the old SVG foreignObject workaround. So it works on Chrome, Firefox, and Safari, of course! #WebGL #VFXJS #Codepen

𝘼𝙈𝘼𝙂𝙄

52,893 views • 6 months ago

The new ASI1-fast just dropped.🔥 We put it next to another LLM. Same prompt. One blazed through. The other model? Still loading. Toggle ‘asi1-fast’, and try it now on Real footage. No skips. 👇

The new ASI1-fast just dropped.🔥 We put it next to another LLM. Same prompt. One blazed through. The other model? Still loading. Toggle ‘asi1-fast’, and try it now on Real footage. No skips. 👇

Fetch.ai

20,547 views • 1 year ago

You can ask Opus 5.5 to write extreme music theory wankery and it will do it in javascript. Theory-porn as a genre? everything you see and hear is generated from javascript code that claude wrote. no samples, no libraries 🔊

You can ask Opus 5.5 to write extreme music theory wankery and it will do it in javascript. Theory-porn as a genre? everything you see and hear is generated from javascript code that claude wrote. no samples, no libraries 🔊

dadabots

44,264 views • 1 month ago

Only the ones that adapt make it to the next stage.

Only the ones that adapt make it to the next stage.

SurrexLabs

126,310 views • 1 month ago

Gemini Spark can now intelligently track topics and react to events in real time. Try the prompt in the next post to get tailored game analyses emailed to you after your favorite team plays:

Gemini Spark can now intelligently track topics and react to events in real time. Try the prompt in the next post to get tailored game analyses emailed to you after your favorite team plays:

Google Gemini

141,485 views • 3 months ago

A Rive customer asked us how to convert their 3D files to vector. So I built this demo to show you don't have to. Instead of baking the 3D into a frame-by-frame vector sequence (the way they'd done it for Lottie in the past), you can keep the real 3D and just make it look vector. The 3D models are standard glTF files, loaded straight into Rive and rendered on the GPU Canvas with flat three-tone shading and outlines. It looks like illustration, but it's still 3D: you can spin it, animate it and relight it, all in real time. The phone screen is a live Rive artboard with real components, layouts, text, data binding and state machines, mapped onto the 3D phone. You can use it in 3D: scroll the app, tap the tabs, it all just works. I built it with the Rive CLI, then pushed it to the Rive Editor, where all of Rive's interactive design and animation tools are there to keep iterating on it. Oh and the whole thing is 344 KB.

A Rive customer asked us how to convert their 3D files to vector. So I built this demo to show you don't have to. Instead of baking the 3D into a frame-by-frame vector sequence (the way they'd done it for Lottie in the past), you can keep the real 3D and just make it look vector. The 3D models are standard glTF files, loaded straight into Rive and rendered on the GPU Canvas with flat three-tone shading and outlines. It looks like illustration, but it's still 3D: you can spin it, animate it and relight it, all in real time. The phone screen is a live Rive artboard with real components, layouts, text, data binding and state machines, mapped onto the 3D phone. You can use it in 3D: scroll the app, tap the tabs, it all just works. I built it with the Rive CLI, then pushed it to the Rive Editor, where all of Rive's interactive design and animation tools are there to keep iterating on it. Oh and the whole thing is 344 KB.

Guido Rosso

11,410 views • 26 days ago

addicted to using boomerang mode (aka reverse D-Mail) in Pi these days. ctrl+alt+b enables it for the next prompt submitted -> after the prompt runs, it rewinds back to the same point with file changes intact + leaves a summary in the feed so the agent knows what happened. using it often can make the context window feel nearly unlimited. Powered by the native /tree functionality in pi. pi install pi-boomerang

addicted to using boomerang mode (aka reverse D-Mail) in Pi these days. ctrl+alt+b enables it for the next prompt submitted -> after the prompt runs, it rewinds back to the same point with file changes intact + leaves a summary in the feed so the agent knows what happened. using it often can make the context window feel nearly unlimited. Powered by the native /tree functionality in pi. pi install pi-boomerang

Nico Bailon

38,562 views • 6 months ago

Introducing HijackedAI . com create an agent - give it sys prompt, a money to hold, and a deadline People can pay to try and convince it to send them the money. Failed attempts stacks in the pool till time’s up. No one broke it - you win prompt wizards, it's time to prove it

Introducing HijackedAI . com create an agent - give it sys prompt, a money to hold, and a deadline People can pay to try and convince it to send them the money. Failed attempts stacks in the pool till time’s up. No one broke it - you win prompt wizards, it's time to prove it

Adam

32,861 views • 1 year ago

This was a song I was writing using SMW samples, planning to later adapt it to the SNES chip in Furnace Tracker. I can say it’s finished, though personally I feel it still needs some small details polished

This was a song I was writing using SMW samples, planning to later adapt it to the SNES chip in Furnace Tracker. I can say it’s finished, though personally I feel it still needs some small details polished

Yiter

17,030 views • 11 months ago

I love this video so much and the one thing I think every time it pops up on my feed is; who the fuck has a bin in the living room next to the telly?

I love this video so much and the one thing I think every time it pops up on my feed is; who the fuck has a bin in the living room next to the telly?

Adam Rowe

219,660 views • 1 year ago

Your coding agent can make animations like this now. I released the skill for drawing them in JavaScript, with 4 visual styles and examples to start from. This 1,5 minute film was made using it. Pick a subject and try it:

Your coding agent can make animations like this now. I released the skill for drawing them in JavaScript, with 4 visual styles and examples to start from. This 1,5 minute film was made using it. Pick a subject and try it:

Alexey Fateev

48,967 views • 1 month ago

Make Stitch designs real with AI Studio From design to a live app in under 60 seconds. Here's how you can now take your Stitch designs and make them real with Google AI Studio: 1. Design in Stitch. 2. Export the HTML. 3. Prompt AI Studio to convert it into a working app. And the best part? This is the manual way. We're working on making this a 1-click experience. Consider this the warm-up.👟

Make Stitch designs real with AI Studio From design to a live app in under 60 seconds. Here's how you can now take your Stitch designs and make them real with Google AI Studio: 1. Design in Stitch. 2. Export the HTML. 3. Prompt AI Studio to convert it into a working app. And the best part? This is the manual way. We're working on making this a 1-click experience. Consider this the warm-up.👟

Stitch by Google

42,480 views • 1 year ago

Using Claude in Chrome product. Prompt: replicate this image (my profile picture) in jspaint, use brushes to replicate it as closely as possible. I want you to use brushes and clicking. Do not use javascript tool, but the final result has to be very close to this image. DO NOT CHEAT 1. I specified the do not use javascript tool because otherwise it uses javascript and it is faster but it seems like people cared more about clicking :D 2. The task took 45 minutes, but I don't think it's that much slower than competitors 😄 3. Prompt is important. (See thread)

Kiana Ehsani

410,934 views • 1 month ago

AI is getting incredibly good. The problem is it still gets stuck whenever it needs to interact with the real world. Need to host a website? API keys. Need to send emails? Another service. Need to use external tools? More accounts and more setup. That's exactly why I wanted to try Zero. Zero gives Claude and ChatGPT access to **14,000+ real-world services**, so instead of stopping to ask for integrations, they can actually complete real tasks. I tested it by asking Cursor to build an AI healthcare startup. Within a single workflow it: → built and hosted a complete website → generated the assets it needed → emailed me the live URL → showed the total cost of the entire workflow No API keys. No juggling different services. No complicated setup. Just a prompt → a finished result. If you're building with Claude or ChatGPT, it's definitely worth trying. The first 10,000 users can claim $5 in free credits to get started: Try out now: #Ad

AI is getting incredibly good. The problem is it still gets stuck whenever it needs to interact with the real world. Need to host a website? API keys. Need to send emails? Another service. Need to use external tools? More accounts and more setup. That's exactly why I wanted to try Zero. Zero gives Claude and ChatGPT access to 14,000+ real-world services, so instead of stopping to ask for integrations, they can actually complete real tasks. I tested it by asking Cursor to build an AI healthcare startup. Within a single workflow it: → built and hosted a complete website → generated the assets it needed → emailed me the live URL → showed the total cost of the entire workflow No API keys. No juggling different services. No complicated setup. Just a prompt → a finished result. If you're building with Claude or ChatGPT, it's definitely worth trying. The first 10,000 users can claim $5 in free credits to get started: Try out now: #Ad

Abdullah AI

12,775 views • 3 months ago

I love the rise of free UI libraries built by designers who really care about the details. Beautiful UI has 17 polished patterns for AI-native products, from thinking and streaming states to tool calls, approval cards, and diffs. Every component includes the source code, so you can give it to your agent and adapt the interaction to your own product.

I love the rise of free UI libraries built by designers who really care about the details. Beautiful UI has 17 polished patterns for AI-native products, from thinking and streaming states to tool calls, approval cards, and diffs. Every component includes the source code, so you can give it to your agent and adapt the interaction to your own product.

Meng To

152,591 views • 3 months ago

MUSE FEATURE ALERT: you can video chat with your Muse coming soon! it talks back in real time, and you get to prompt exactly how it sounds! (iphone screen record doesn’t record the sound so you’ll have to try it yourself with your own muse to hear the voice)

MUSE FEATURE ALERT: you can video chat with your Muse coming soon! it talks back in real time, and you get to prompt exactly how it sounds! (iphone screen record doesn’t record the sound so you’ll have to try it yourself with your own muse to hear the voice)

Alexandr Wang

149,355 views • 1 month ago