Six months ago, my design process started in Figma. I'd open a blank canvas, create some frames, and start pushing pixels. That workflow served me well for years.
It's not how I start anymore.
Now I start with a conversation. I describe what I'm thinking to Claude Code, we iterate on the structure together, and within minutes I'm looking at a working prototype in a browser. Not a mockup. Not a static screen. A real thing I can click through, test, and share.
The feedback loop collapsed
The biggest shift isn't the tools themselves. It's what happens to the feedback loop when you can go from idea to working prototype in an afternoon instead of a sprint. You stop treating exploration as expensive. You try more things. You throw away more things. The ratio of ideas explored to ideas shipped goes up dramatically.
This changes how you think about design decisions. When a prototype takes two weeks, you're incentivized to get it right the first time. When it takes two hours, you're incentivized to get it in front of someone and learn. The economics of experimentation just changed.
Tools I didn't expect to love
Git was the tool I resisted longest. Version control sounded like an engineering problem. But once I understood it, I realized designers have been solving the same problem badly for years. Git just does it properly. And deploying through Vercel means I can push changes and see them live in seconds. That immediacy changes how you iterate.
Then there are the tools built for this moment. Design tools that sit closer to code than to static mockups. Platforms where someone can drop a comment directly on a live site and an AI agent handles the change. The distance between feedback and implementation is approaching zero.
I'm using Obsidian as a thinking tool, connecting notes and references in ways that feed into design decisions. And Claude Code has become less of a coding assistant and more of a design partner: something I think with, not just build with.
What this means for product design
I think we're at an inflection point. The role of a product designer isn't shrinking. If anything, it's expanding. But the skills that define it are shifting. Taste still matters. Systems thinking still matters. Understanding people still matters. What's changing is the execution layer: the distance between an idea and something someone can experience.
Designers who lean into this will be able to test more ideas, move faster through ambiguity, and collaborate with engineers in their own environment. Designers who resist it won't become obsolete, but they'll be working with a narrower set of tools than the problems demand.
I don't know exactly where this goes. But I know my process looks nothing like it did six months ago, and I'm designing better work because of it. That feels like something worth paying attention to.