Some of the best designers I’ve ever worked with all had one thing in common—the ability to communicate their design decisions into words. It’s a very difficult task. It feels like an absolute black art to me. In the same way that maybe coding feels to someone else. But the designers that are able to translate are the ones that are going to come out on top in this AI world. How can you teach someone to design like you? To think like you? When you work on a design system, how much of “you” is allowed to permeate through? How much of your work is just playing with someone else’s toys (RE: LEGO)? Isn’t that the goal of a design system?
In fact, this translation has previously been so difficult that we’ve conceded to compartmentalizing design decisions into componentized shipping containers. “Atomic design” begat components begat design tokens. Each layer gets farther from the truth, that the reason that your line height is set to 1.1 is because your application is, or was at one point, very data-intensive and thus you needed to optimize for information density. Because one time someone complained about not being able to see a very important row in a table and that mistake cost so much money that you were hired to redesign the whole system. But that’s a mouthful. You can’t throw that over the wall. An engineer can’t implement that. So we make little boxes with all batteries included.
Except now things have changed and we can throw all of that over the wall. Every decision. Every vibe. We no longer need to compress our thinking into static artifacts. LLMs give us the ability to ship our exact train of thought, uncompressed, a little bit lossy but still significantly useful. Full context that is instantly digestable. Instead of shipping <Boxes>, ship a factory.
You can teach an LLM to think like you. To design like you. Some of you might feel threatened by this idea. Maybe you feel like it reduces you or makes you irrelevant. A machine can never design. To those that feel this way, I offer no persuasion.
But to those that remain curious, once you manage to figure out how to tame the machine, man, I can’t imagine a greater relevance. Your perspective imbued across every component, pamphlet, and pixel. A voice at every table.
Start today. Download Claude Code. Tell it to design something. When it gets it wrong, translate. Steer it. Create waymarks. Express your intent. What isn’t working? When you’ve made an inch of progress, throw it all away and start over. Over time, build your personal pattern language. Save it as a text file. Enter it before every prompt. Bring it to your next feature, project, job.
And if this whole AI thing turns out to burst, at least you’ve improved the one skill that some of the best designers I’ve ever worked with had in common—the ability to communicate their design decisions into words.