Good design
ships and keeps moving

My workflow is built around one thing: getting from fuzzy problem to clear, working solution, fast and without losing the craft. Every tool, every habit exists to reduce friction between thinking and making.

BIAS TO ACTION

DESIGN SYSTEMS

AI AUGEMENTED

Design Systems

Prototyping

AI-Powered

Web/Mobile Apps

Product Strategy

Landing Pages

UX / UI Consultation

The tools, rhythm, and principles behind how I work

The tools, rhythm, and principles behind how I work

Click on any tool to see when and how it fits.

CLAUDE

Claude + MCPs

CORE WORKFLOW

FIGMA

Figma + Figma Make

DESIGN + PROTOTYPING

LOVABLE

Lovable

FUNCTIONAL BUILDS

Figma

Google Stitch

EARLY UX

Figma

Vercel

DEPLOY + REVIEW

Figma

Granola

NOTE AUTOMATION

The rest of my stack

Trello

ChatGPT

Claude

Figma

Framer

Webflow

Rive

Blender

Trello

ChatGPT

Claude

Trello

Webflow

Claude

ChatGPT

Rive

Claude

Blender

Figma

Trello

Framer

ChatGPT

01

A week in my workflow

Not every week looks like this but most do

MON

Align & plan

Review sprint goals, surface Granola notes from last week's sessions, sync with the PM. Use an LLM to draft a quick brief and surface risks before the week starts.

CLAUDE

GRANOLA

TUE

Concept & structure

Stitch for IA sketches and early flows. Sometimes generate an HTML preview to align with engineering before I open Figma. Gets the interaction model right first.

STITCH

CLAUDE

WED

Design & system work

Deep Figma sessions. MCP integration keeps the design system healthy: token audits, component states, pattern consistency throughout.

FIGMA

CLAUDE MCPs

THU

Prototype & validate

Interactive concepts and quick usability checks. Deploy builds to Vercel for async stakeholder review. No scheduling required.

FIGMA MAKE

LOVABLE

CLAUDE

VERCEL

FRI

Handoff & contribute

AI packages dev handoff specs with zero back-and-forth. Share patterns and prompts with the wider team. Retro notes captured via Granola.

CLAUDE

VERCEL

GRANOLA

01

How I think about work

Guidelines I work by and my non-negotiables

01

Own it end to end

Partner with PM and engineering from problem definition through launch and iteration, not just at the handoff gate.

02

Start in the LLM, not the canvas

Clarify intent, surface edge cases, draft a short brief. Align the team before a single frame is drawn.

03

Research as velocity, not a gate

Talk to customers, run quick tests, adjust direction. Treat validation as a speed tool, not a process tax.

04

Design for the 80/20

Encode judgment and complexity under the hood. Keep the default experience simple and successful for most people.

05

Contribute to patterns, not just pixels

Share prompts, patterns, and learnings. Participate in crits that raise the bar for the whole team.

06

Build the system before you need it

Token audits, component states, and naming conventions feel like overhead until three people are touching the same file.

07

Treat AI as a draft partner

Claude and MCPs accelerate the work. The final call is mine, and so is the responsibility for what ships.