That process works but the gap between "here's what I think the product should do" and "here's proof that it works" used to take a long time - weeks! Wireframes are usually static. They communicate layout and hierarchy, but they can't answer questions like "does this flow actually make sense when you click through it" or "what happens when the API returns 47 results instead of 3."

Over the past year I've built a workflow that closes that gap. I go from a PRD and user journeys to interactive, coded wireframes in hours. The tools are Claude Code, Cursor, and pre-built components.

How it works

The process has four stages.

Research and PRD. I start the same way I always have. User research, business goals, constraints, jobs to be done. I write a PRD that captures all of this. The difference is where it goes next.

User flows. I feed the PRD into Claude Code and ask it to generate user flows based on the goals and user needs. These are flows I can put in front of stakeholders, product managers, and engineers to validate before any UI gets built. We align on the logic of the product before anyone talks about what it looks like.

Coded wireframes. Once the flows are validated, I feed them back into Claude Code alongside a brief covering design and engineering requirements. It generates interactive wireframes using shadcn components. These are functional. You can click through them. You can test edge cases. You can hand a laptop to an engineer and ask "is this feasible" and they're looking at real code, not a picture.

Figma for high-fidelity. Once the direction is validated and engineering has confirmed feasibility, I move to Figma for the polish layer. Component refinement, visual design, brand expression, design system alignment. The high-fidelity work happens with confidence because the underlying flows have already been tested.

Why I believe this order matters

Getting buy-in is one of the hardest parts of the design process. Engineers want to know if something is feasible. PMs want to know if the flows cover the right user needs. Founders want to see the vision come alive before committing resources.

Static wireframes make all of those conversations harder. You're asking people to imagine how something works based on a series of rectangles. A coded prototype lets engineers assess feasibility on real code. It lets PMs click through the actual user journey and flag gaps. It lets founders experience the product direction instead of interpreting a deck.

By the time I open Figma, buy-in already exists. The flows have been validated by the people who need to approve them, using something they could actually interact with. The high-fidelity design phase becomes about craft and detail because the strategic decisions have already been made with confidence.

Existing products and design systems

This compresses even further when the team has a design system in code. You're prototyping with the same components that ship to production and the gap between prototype and shipped product gets very small.

I can explore a new feature direction in an afternoon, test it internally, get buy-in from engineers, PMs, and founders, and hand off code that engineers can use as a reference for the production build. The conversation around feasibility and direction happens early, with working code in front of everyone.

What stays the same

The designer's job here is the same as it's always been: understand the user, understand the business, and make sure the product serves both. The research phase is unchanged. The thinking is unchanged. The strategy is unchanged.

What changed is how quickly I can turn that thinking into something testable. And how much earlier I can involve engineers and stakeholders in the conversation with something real in front of them.

Here's the link to the current project I'm working on and the current state of the tool: https://starlit-bienenstitch-5611b8.netlify.app/

Louis is a product designer with 10+ years of experience building financial products at MetaLab, Kraken, and 10M Studio. He's the founder of OpenHouse, a platform making UK property investment accessible from $20.

Vignelli spent his career proving that complex systems could be made legible to anyone if you cared enough to do the work. The NYC subway. American Airlines. Knoll. IBM.

"If you can design one thing, you can design everything,"

He said this because the principles are universal and they transfer across mediums, industries, and eras.

Those principles are over fifty years old and ready to be applied to crypto because it's is ready for that same clarity.

Crypto UX does seem to be improving through embedded wallets, account abstraction, gas fee abstraction, and fiat onramps.

The data shows where the biggest opportunities still live: only 13% of Americans find crypto wallets easy to use, up to 90% of exchange users abandon before their first transaction, and only 13% of wallet users return after their first week compared to over 60% for traditional fintech apps.

These are design opportunities ripe for the taking.

The Principles Worth Applying

The greatest designers of the last century shared a set of beliefs that went deeper than usability checklists, and those beliefs are exactly what crypto needs now.

• Emotional clarity

• Radical reduction

• Material honesty

• Choreography over layout

• Respect as a design principle

Jony Ive, Dieter Rams, and Vignelli all understood that great design is a form of care. Ive designed objects that dissolved into behaviour, Rams removed everything until only the essential remained, and Vignelli made the most complex systems in the world feel obvious. The crypto products that apply these principles will be the ones people actually use, because when someone swaps a token or funds a wallet or bridges between chains, the experience should feel as natural and considered as turning a page.

Who's Getting It Right

@worldcoin

Strips the interface down to only what's useful in everyday life: gas-free transactions, an 18MB app size, support for 98% of Android devices worldwide, and the blockchain running underneath without the user ever needing to know.

@summerfinance_

Handles complexity with transparency through clear risk visuals, one-click leverage via flashloan-powered multiply features, and stop-loss automations that give users safety nets. Their CEO Chris Bradbury has said his mission is to make self-custodial finance as usable as it is powerful.

@aave

Rebuilt their entire interface for V3 with fresh eyes, giving you net worth, supplies, and borrows in one clean view with information cards that educate users within the flow.

@Uniswap

Understands that a swap is a swap, and the interface reflects that with a single clear action, beautiful UI, and the complexity of routing across liquidity pools handled entirely behind the scenes.

@fusewallet

Takes the same approach to everyday payments, giving users a clean and focused experience that solves a specific need effectively without asking them to understand the infrastructure underneath.

What these products share is a mindset of asking "What does this person need to accomplish?" and building from there.

I'd love to add the Base app to this list, but the experience still feels like it's trying to achieve too many things at once and as a user it leaves me unsure of what I'm actually there to do.

What's Next?

60% of users drop off before funding their wallet, 85% who connect to a dApp take no further action, and over 70% leave after a single transaction. In any other industry, numbers like these would trigger an all-hands product review, and crypto has the same opportunity.

Don Norman wrote in The Design of Everyday Things that when things go wrong, people blame themselves, but every moment of confusion is really a design opportunity telling you exactly where the experience can improve.

The next hundred million crypto users will come from products designed by people who actually care about the experience, tested with real users who have never touched a wallet, built by teams willing to ask: would my mum understand this?

Vignelli proved that the most complex systems in the world can be made clear with enough discipline, because a city of eight million people navigates 472 stations across 36 lines every day thanks to one designer who committed to making clarity the priority.

Worthy reading

• The Design of Everyday Things, Don Norman.

• Don't Make Me Think, Steve Krug.

• Ruined by Design, Mike Monteiro.

• About Face, Alan Cooper.

• Refactoring UI, Adam Wathan & Steve Schoger.

Louis is a product designer with 10+ years of experience building financial products at MetaLab, Kraken, and now 10M.

He's the founder of @Openhousefi, a platform making UK property investment accessible from $20.