Your job haunting, sorry, hunting shouldn't be boring 🫣 So, i built a better hiring platform where you are haunted, sorry, wanted person on a 3D interactive brick wall developed in @threejs using Claude Fable 5 It was super fast to built, probably, a less than hour of build with fun little experiments. Everything that you see here is a full WebGL scene with ACES filmic tone mapping. That whole brick wall is code with an irregular per-brick polygon outlines, fired-clay HSL, kiln streaks, cracks, edge chips, efflorescence, grit, plus a matching bump map with recessed mortar All of it, AI did it on its own. I just had a trash talk with it on what has to be built. Wouldn't it be super cool if companies adopt it and it becomes more like a nice experience of feeling where it feels like a 90's real street corner Live: Code:

Been clauding so hard these days that @PacktPublishing decided to send me this Claude Skills book by @PietroMontaldo so I could finally push myself to turn my workflow into actual skills haha 🤭 Would you be interested in having my workflow available as skills? Maybe one skill to help you write better prompts for AI, another to generate assets, another to turn those assets into 3D models, and finally one to ship everything into a website. Or maybe this entire process should just be one big skill 👀 Let me know what you would prefer to learn and actually use. Also, thank you @Prajwalyes for sending this over. Really excited to go through the book and start creating some Claude skills 🧡

One of the best things about AI right now is how easy it is to follow a random idea 🔥 I saw a PlayStation related post on X and got an idea for a game controller exploration website. I searched around, found a design on Pinterest that I liked, and decided to build it. I turned the design into visuals with GPT Image 2.0, created the 3D assets with Tripo, and then gave everything to Claude Opus 5 to turn it into the actual website you see here. That simple. And the best part is that it didn't even take that long. Just a couple of hours of focused effort. The interesting thing isn't that AI made a website for me. Although, yes, that's pretty cool too. What's more interesting is that I could have a random idea, explore it, build the assets, and turn the whole thing into a working website within a couple of hours. That's what makes AI so exciting. It wasn't so possible before. Live: Code: Design:

Tbh, Gemini 3.7 Flash still isn't there yet 🥶 I don't see any meaningful improvement over Gemini 3.1 Pro or the other Gemini models. I tried vibe coding a dashboard in Google AI Studio. It got some things right, but constantly failed at others. I repeatedly asked it to redo sections to match the design, but it just wouldn't. No visual changes at all. Although its speed is impressive compared to Kimi K3 and even Claude, it lacks intelligence to create things the way user requests. This has been a problem with Gemini for a long time. I've built quite a few apps with it, and while it can get things done with a few more iterations than other models, eventually it just stops following requests. You can keep asking. It simply doesn't make the changes. Google really needs to step up its game. Studio: Design:

I genuinely don't have any words for this 🤯 Just ONE image of a robotic bee and somehow got this incredibly gooood 3D model with great topology. No prompt. No rework. No retries. Nothing else. This kind of thing could have taken weeks of modeling, retopology and iteration. Or maybe more. Damnn Now anyone with a crazy enough idea can just make beautiful things. Anyone. That honestly feels insane. What a time to be alive!!

How many of you actually know how "what you eat" grows? 🤔 I never learned crop science in school. I learned it by walking around our farm with my father, who showed me different crops and explained how they grow. But not everyone has access to a farm, so I built a 3D Crop Science application in @threejs using Claude Opus 5 to make that learning experience more visual and interactive. It is an exploration of different types of seeds where each seed has three different stages. The first stage shows the whole seed. The second stage is a cross section that shows what is inside the seed and how it grows. The third stage shows the germination phase through a 3D model. There are 24 models in total. The original models were 1,349.6 MB, which Claude optimized down to just 11.6 MB in total. All of the 3D models were generated using @tripoai. Claude also created visual information for each seed showing details like seed type, storage tissue and germination phase. Each seed also has a YouTube video that shows how it actually grows in the real world. There is also a section showing where each seed is commonly grown using both a 3D globe and a flat map. You can explore different regions and click on specific areas to see more information about where that seed is grown. The entire experience is fully responsive and also includes quizzes that you can take in both normal and 3D formats. I strongly believe education becomes much more effective when you can see, interact, watch and explore instead of relying only on textbooks. Live: Code:

Lovable might have captured a much bigger market than we think. Saw an author, who has around 85k followers, promoting their book website on an Instagram channel which is built using Lovable Kind of wild how far Lovable has reached beyond the usual tech crowd. Authors and publishers are vibe coding now too. Soon there will be so many more cases like this Love seeing this. This is the future we are heading towards Just wish either the author knew SEO or Lovable helped them somehow with intelligence lol 😂

It still surprises me how Claude turned a static model into a moving animated one without touching Blender or using any MCP I checked how it actually did this by digging into its thinking process It wrote code that reads the real coordinates inside the file, checks the pose is right, and renders images to confirm before changing anything. What an intelligence is that! Though we're still early here, 3d modeling is truly about to change a lot Tbh, a future where AI builds a 3d model and animates it just from a plain English words isn't far off

This is for everyone who still thinks Gen3D is useless 🤣 I generated a koi fish with clean topology using @tripoai but it was just a static mesh. No rig. No bones. Nothing. I wanted to animate it without touching Blender because I genuinely have no idea how to use Blender 😂 So I casually asked Claude if it could help me with it... And then Claude built THIS 🐟 A fish that swims, breathes, reacts to your cursor and even looks directly at you when you stop moving your cursor. It took a completely static 3D model and turned it into a fully animated fish with a swim cycle and subcarangiform kinematics. It even optimized the model and brought the file size down from 14.79MB to just 4.16MB. And also that water with ripple effects is built with TSL too. 60 FPS. WebGPU with WebGL2 fallback. All @threejs. This was inspired by @marina_uiux's website where I saw a fish swimming across the page. I immediately loved the idea and thought, "What if the fish actually felt alive, more immersive?" So I made it happen. And honestly, this is the kind of thing that makes me excited about what we can build with 3D on the web now. All thanks to AI! So for everyone who thinks you need to spend months learning Blender before you can build something like this, maybe it's time to rethink what Gen3D or AI is actually capable of :D Live: Code:

Thought I’d let my 220 butterflies say hello to you in 34 different languages using @threejs 🦋 Built using Claude Opus 5. It’s a little fun project where a cute lil’ butterfly follows your cursor and disturbs the rest of the swarm as you move closer to them. And yes, it's all TSL. Live: Code:

Are we going to say bye bye to Blender in the future? We’re still REALLY early in AI 3D modeling, yet AI can already generate surprisingly clean 3D topology from a single image. It’s far from perfect, but the direction is pretty insane. And this is probably just the beginning, isn't it? 👀

Grok 4.6 spent 194K tokens trying to turn this CD player concept into a mobile app 😈 I really loved this CD player concept so I wanted to see how well Grok could turn it into an actual react native app I asked Claude, which generated the original CD player website, to create a detailed prompt describing the UI, screens, interactions, and everything that had been built, and then gave that prompt to Grok. The result is okayish at best, or maybe just bad 😅 As you can see, it’s quite laggy and not very well executed, though this was a one shot attempt with no iterations or back and forth I only gave Grok text instructions and no design image, so that’s probably why it didn’t do very well in terms of the design (though it had whole design system explained in provided prompt) From my fun experiments, I’ve noticed that models seem to do much better when they’re provided with reference design images But that’s just the design part of thing. I was hoping Grok would at least handle the logical part well, especially the whole CD dragging animation natively It kinda failed there 😅

AI can build beautiful animations if you describe what you want clearly enough ❤️‍🔥 Built this CD player with smooth subtle animations inspired by @itsdpark and @joshpuckett’s concept using Claude Opus 5. All I gave Claude was the design image and my master prompt describing how the animations should work. I explained things like how the CD should be dragged onto the player and replace the currently playing CD, or how it should subtly increase in size while being dragged. I didn’t use any technical animation terms or talk about easing feasing at all. I just used simple words to describe what I could see happening in the video. Simple. That’s really all you need to build beautiful animations with AI! Live: Code:

If your emails were from royal times 👑 Built a cool cinematic 3D letter inbox in @threejs using Claude Opus 5. Everything you see is pure code, no external assets. Even the leather wax seal was built using LatheGeometry. I honestly didn’t expect AI to pull this off so well. The shaders, noise effects, rods, and especially the royal writing turned out so beautifully. The text specifically feels like it’s soaked into the paper rather than printed on top because it’s composited directly inside the shader. AI made that decision. Really happy with how this turned out. It was such a fun build during some downtime. Vibe coding makes it kind of crazy how quickly you can go from a concept in your head to something that actually feels real, right? Live: Code:

You don't need to be a 3D artist to start building 3D websites anymore 💛 Built this cool 3D jet drone website in @threejs using Claude Opus 5 and @tripoai in < 1 hour. A lot of people still think AI is great for generating code, but not for building something that actually feels polished and interactive or "complex". But it has been changing super fast these days. The model you see here is only 946KB, compressed from its original 13.6MB size automatically by Claude without losing any quality. This kind of intelligence has been developed in many other AI tools for anything related to 3D (except GPT 5.6 sol lol). You just have to describe what you want and they handle everything "engineerly" (that's not even a word but okay :D) Anyway, also you don't really have to learn anything about 3D modeling to create these models. All you need is a good image and put it in Tripo. For this case, i generated model using "smart mesh" feature that generated a cleaner low-poly topology and edge flow. That helps keep your actual 3D asset even more lightweight. Earlier, with HD model feature, my model size with 8k texture was around 60-70MB but with Smart Mesh, it's around 13-16MB with 8k texture applied separately. Of course, this isn't something I knew from the beginning. All I did was "try" out different features and see what happens. Live: Code: Design: