AI has made it much easier for designers to become builders. It has not made it equally easy for builders to become designers.

A snapshot, a contract per component, and a check that catches drift, unbound tokens, and broken variants — no MCP server needed.

ANE systems emphasize environmental determinism, structural consistency, and rapid validation to allow agents to produce production-grade changes with minimal human oversight.

Claude Cowork is the first tool from Anthropic that treats knowledge work the way we designers actually experience it: as multi-step tasks with real files, real deliverables, and a chain of judgment calls in the middle.

Despite the genuine loss associated with the squircle restriction, there's more than one way to design with it.

The final interface lives in code, which is where it should live – we are software designers, after all.

A survey of 604 tech and design professionals found that “UX” remains the default name for our field, while alternative names remain fragmented.

Relying on the same combination of product quality, SEO, launch platforms, and occasional social posts feels increasingly naive.

The leap from that painted stage to a world you move a camera through is the same leap CSS lets you make, and it's smaller than it looks.

The line height unit (lh) has been around for a while and I rarely see people sharing how they use it.

AI did not change who we create for, it changed who reads first.

What’s interesting and I bet the main reason Asana approached it this way, is that Tab is a separate little island, far away from other modifier keys.

That feedback loop between imagining, seeing, and building has become dramatically shorter. And suddenly, a whole world of ideas is within reach.

If you want the resources necessary to build something well, then choose to work on things worth doing well.

If there's no result that could prove you wrong, you haven't actually said anything.

An agent loop can pass every verifying test and still make a product worse.

The sad gray icons looked like a taste problem, but the biggest improvement came from giving the agent access to product judgment I had already worked out.

The question to sit with: A year from now, when everyone on your team has the same tools and the same prompts and skill files, can anyone tell your work apart from theirs?

Taste is the work after generation, when you read the output, name what's wrong and change it before anyone else has to live with the draft.

As AI becomes increasingly powerful at generating ideas, our judgement of those ideas becomes increasingly important.

We do it, naturally, for you, as we set out to rebuild our online editor to better support what today's developers want and expect.

Why "systems thinking" is now the most important skill she looks for.

TJ explains why sprinkling AI into a design system without proper context creates chaos rather than speed.

Strip away the branding, and nearly every screen you used today was assembled from roughly a dozen standard parts: a button here, an input field there, a menu across the top, links in the middle, and a dialog box interrupting at the worst moment.

Generation won't tell you when to stop looking, taste is the stopping rule.

Interface Lineage takes them apart for the people who make them. You leave able to name each move, feel how it works, and say why it matters.

It all makes me wonder what other aesthetics are being born out of this AI moment and how many will spread, take seed, and become part of common software interaction paradigms for years or decades to come?