For over a decade, I followed the same process:

Research → Wireframe → Design → Prototype → Handoff

It felt responsible. Thorough. Professional. But it was built for a world where building was expensive.

Building isn't expensive anymore.

Six months ago, I started using Claude Code. I'm not a developer. I don't write production code. But I can describe a component in plain English and have something running in my browser in minutes.

Not polished. Not shippable. But real enough to react to.

That changed everything.

The old process assumed you needed to think before you made.

You'd spend days in Figma getting the spacing right, the states covered, the flows mapped — all before anyone clicked on anything real.

Then you'd hand it off. Wait. Get feedback. Realize half your assumptions were wrong. Go back. Repeat.

The process protected us from wasting engineering time. But it wasted something else: clarity.

The new process is backwards.

Now I start with a conversation. I describe the problem to Claude. It builds something rough. I click around. I notice what feels off. I find the states I hadn't considered.

Then I open Figma.

Not to explore. To refine.

What dies:

• Designing screens before understanding constraints

• Pixel-perfecting things that might not ship

• Writing specs that get misinterpreted

• Weeks of work before real feedback

What lives:

• Thinking through making

• Validating ideas in hours, not weeks

• Designing with technical constraints visible

• Shipping faster, learning faster

  
  

Design isn't dead.

The craft still matters. Taste still matters. Knowing what to build and why — that's more important than ever.

But the linear, waterfall-ish process we inherited? That's dying.

The designers who adapt will move faster, learn faster, and have more impact.

The ones who don't will spend weeks polishing things that never ship.

The tools are here.

Claude Code. Cursor. V0. Figma MCP.

You don't need to become a developer. You need to become comfortable building rough things fast.

The prototype isn't the deliverable. It's where you find out what you actually think.

I'm documenting this shift as it happens. Follow along if you're figuring it out too.