A thread on how Zen Garden came to be. Built completely with Opus 5.5. 25 billion tokens later 🧵 It all started about 2 weeks ago when @Nalden shared @gwern's essay on samon, the patterns raked into the sand of zen gardens. I got hooked right away with the story behind it. Being a Codex stan, I wanted to try out Opus 5.5 see what all the buzz was about, so I gave it a simple prompt to get a quick vibe check to see if we should pursue it further to turn this into our next app. In mere minutes I had a working prototype. It was fine. It was very flat. Nothing special. Then something happened that I hadn't felt in almost four years of building apps with AI. I simply asked Claude to make the scenery more realistic. It started working for a while, then it came back and my jaw dropped. Given how good these models are now at using Blender, I sorta expected it to create 3D models. Instead it told me: "There are no images, textures or 3D models anywhere in this. Every pixel is calculated from scratch by maths, every frame." Moss, rocks, water, cherry blossoms. All of it just straight up code. Metal shaders. I would not have even considered that as an option. Which became the pattern of my experience building Zen Garden with Opus 5.5. Taking the unexpected harder route to get to a better result. No shortcuts. So when @michelelings suggested "would be cool if a bird came flying into the garden." Opus came back with the birds that actually visit Japanese gardens, life-size, each with its own call sounds. It calculated where it could land, its motions. It opened a garden on my screen and a kingfisher landed on the bridge, watched the koi, and dove in the pond to catch a little fish. I f***ing lost it. I didn't ask for that. I just asked to 'add a bird'. Every time I did ask for something, it did it in a way I simply didn't consider it would be able to do at the level and in the way it did. Better than what I had in mind. It changed how I now build. Giving me a different perspective of what an app should feel like when made by people who care about every detail. Just asking it to do more, go even harder on the tiniest details. Just because you can. For the love of the game. In June, after we won the Apple Design Award with grug, I wrote that AI can't care, and that you have to bring the taste. I still believe that, mostly. But this was the first time the model actually kept pushing me to be more creative, in a different way than before, challenging me to think much bigger. Crazy how much can change in a just few months. This new way of building gives me so much energy all over again to build more and more. The new outcomes that are now possible. Bigger. Bolder. More creative. For designers like me, Opus 5.5 is the best model out there by a long stretch. And here is the crazy part. With how fast AI is progressing, it will be the worst model in a year. Think about that. But more importantly: plan ahead for that to become reality. Our process to build our apps keeps changing every day, learning new things. Redefining what a "designer" is day to day. And with the latest models it really allows us to be even more creative, more ambitious. "Shoot for the moon. Even if you miss, you'll land among the stars." Two weeks. Twenty-five billion tokens. Zen Garden started as a quick burst of inspiration, but shipped as a piece of art. Thanks to Claude. Receipts below 🧾

grug is probably the first Apple Design Award winner built by two designers using Codex to write the code. We didn't prompt "build me an award-winning app, make no mistakes.” It did not wake up one morning and decide the world needed grug. We did. We wanted to build it and AI was the tool that helped us take it all the way there. Ever since we started building with AI a few years ago, we got bolder. You stop killing ideas just because they sound too hard to build. You get weirder when trying things becomes cheap enough to be silly again. You start following the strange little thought further than you normally would. That is how grug happened. AI can write your code. It can help you move fast. It can make impossible things feel possible. But it cannot care. It cannot make your app memorable. It cannot make your app feel like it has a soul. You have to bring the taste. You have to stay incredibly close. You have to take small steps, make thousands of tiny decisions, throw away good-enough work to get to your best work, and protect the thing that made the idea worth building in the first place. grug would not have been this memorable if we didn't have Codex to go all-in on all the crazy ideas we had. If we had not decided to make the whole thing hand-drawn. If we had not spent days and nights obsessing over every detail, every animation, every interaction, every tiny bit of weirdness. That is the difference between slop and something with a soul. And I think that is why this award means so much to us. Not because what we were able to do using AI to build grug, but because Apple recognized the care we put inside all the weirdness. For the past 15 years, so many of our ideas stopped at the mockup. They were too weird, too small, too hard to explain, too expensive to build, too unlikely to survive a meeting. Now designers like us can build the fun little things. Designers like us can ship the crazy ideas. Designers like us can make unreasonable little things and see if the world cares. There are plenty of people who do not get grug. That is totally fine. We do not try to build for everyone, because when you do you end up building for no-one. The people who get grug really get it. And if it makes their morning feel a little lighter, that is all that matters to us. Make the weird thing. The right people will find it. grug no wait for permission. grug back sun rise.

grug leave cave. grug cross river. cross ocean. sit in flying metal bird many sun. fruit company give grug shiny brick. grug smile whole way home.