Why I Design and Code

There is a gap between design and development. Not a technical gap — a communication gap. Every handoff is a game of telephone. Meaning gets lost. Intent gets diluted. The thing that ships is never quite the thing that was designed.

I decided to close that gap by being both.

The Handoff Problem Is Not New — But It Is Getting Measured

In traditional teams, a designer creates a mockup. A developer interprets it. The interpretation is always an approximation. Padding gets rounded. Animation timing gets simplified. The micro-interactions that made the design feel alive get cut for scope.

This is not anyone's fault. It is a structural problem. When the person who imagines the experience is not the person who builds it, fidelity is the casualty. The breakdowns are temporal, not technical — designers disengage too early, developers get involved too late, feedback loops shrink under deadline pressure. Weak handoffs shape long-term velocity: inconsistent implementations slow future changes, and small updates require disproportionate effort.

Most handoff failures are not loud. They accumulate as friction — deadlines slip from unclarified details, designers notice inconsistencies requiring rework, developers make undiscussed pragmatic trade-offs. The cost is invisible until it compounds.

The Industry Is Moving This Direction

Figma's State of the Designer 2026 report surveyed over 900 designers globally. Eighty-two per cent of design leaders say their need for designers has increased or stayed steady. Seventy-three per cent say AI tool proficiency is now a must-have in candidates. The T-shaped designer — deep in one discipline, capable across others — is the most hireable profile going into 2026.

But the most interesting shift is not about AI tools. It is about designers writing code.

Intercom now has all their designers shipping code to production. Their framework has three tiers: fixing UX debt and content directly in the codebase, vibe-coding interactive prototypes that influence product direction, and owning entire frontend features end-to-end. Designers at Intercom do not need to know the full JavaScript framework or CSS architecture. They have just enough skill to have impact. Atlassian's Rovo Desktop team followed a similar pattern — designers vibe-coding in the real codebase, shipping refinements as pull requests for engineering review.

These are not side experiments. These are design teams at scale restructuring around the idea that the person who designs the thing should be able to build the thing.

The Tools Are Closing the Gap

In February 2026, Figma shipped their MCP server — a protocol that connects design files directly to AI coding assistants like Cursor, Windsurf, and Claude Code. Component properties, variants, and design tokens flow from Figma into the IDE. Developers report saving roughly ninety minutes per week by not switching between the IDE and browser. The initial code scaffold comes with exact floating-point values from the Figma API, mathematically identical to design specs.

The best Figma-to-code tools in 2026 reduce manual frontend development time by thirty to sixty per cent. This does not replace developers. It eliminates the translation layer — the same gap I closed years ago by learning to code.

My Path

I design in Figma and I build in code. When I design an interaction, I already know how I will implement it. When I write a component, I already know how it should feel. There is no translation layer. The intent survives from concept to production.

This does not mean I do everything alone. It means I can move between disciplines without losing context. I can prototype in code instead of static mockups. I can make design decisions in the browser. I can ship faster because I am not waiting for anyone to interpret my intent.

What This Means for the Work

Every project on this site was designed and built by me. The code is the design. The design is the code. They are not separate deliverables — they are the same thing expressed in different mediums.

The industry is validating what I have believed for years: the best digital products come from people who can think across disciplines. The tools are catching up. The org structures are catching up. The question is no longer whether designers should code. It is how much of the stack they should own.