How I work

An ever-evolving design process.

Vision → spec → design → pair-code → dev handoff, with AI co-pilots in the loop end to end — and a research repository the agents read from, not a static deliverable. What used to be a quarter is now a sprint. What used to be a sprint is now a day.

The work has shifted.

Same fundamentals — discover, design, build, ship, measure. What changed is what sits beside me at every phase. The agent does the typing; I do the deciding. The shift isn't toward "AI tools" — it's toward building the agents themselves, and each one I ship makes the next one faster.

Three rules underneath all of it:

R_01

AI as accelerator, not author

The agent does the typing. I do the deciding. Judgment, taste, and accountability stay with the designer.

R_02

Real data over guessed data

Agents read shipped components, real research, real outcomes — never make-believe. The design system is the contract.

R_03

Always shipping, never spectating

Live prototypes, real users, measurable outcomes. The post-launch loop is where the next iteration starts.


Agents around every phase. Conversations stay human.

Most of the product lifecycle now runs through automation grounded in real project data — not AI making things up. The one part that doesn't compress is the conversation itself: sitting with a user, asking the right follow-up, hearing what they don't say. Everything around it compounds.

A_01 · USER RESEARCH

The conversation stays human. Everything around it — automated.

Past studies systemized into a queryable insight repository. Live sessions auto-transcribed and summarized. Insights clustered, scored, and routed into Asana as tagged roadmap items, sized against intent. From there the same loop keeps going: wireframes generated from the synthesis, clickable prototypes coded for the next round of testing, and Figma components plus a living design system auto-generated so stakeholders can comment, mark up, and collaborate where they already work. One tool — research to design to dev. The agent doesn't sit in the conversation, though — that part is still a person in the room.

Built end-to-end inside the AWS UX Research App — research → roadmap → wireframes → prototypes → Figma. Currently in flight.

User testing session — participant working on a laptop, observers taking notes
// observed_session
Jay presenting at a whiteboard, sketching layout ideas with a pen
// whiteboard_sync
Jay walking through dashboards on screen with a colleague
// review_session
A_02 · DESIGN → SHIP → LEARN

From spec to ship to evidence — one continuous loop

  • Spec → wireframes → clickable prototype. Insights turn into a spec (requirements.mddesign.mdtasks.md), wireframes generate on the real design system, and a clickable prototype is in front of users on day one. I review for taste and intent — drift caught at the spec layer, not in production.
  • Pair-coded into production. From there I pair with coding agents (Kiro, Cline, Q Developer) to take the same prototype the rest of the way. Engineers pick up working code, not screenshots — no translation step, nothing lost.
  • Launched staged and instrumented. Success metrics defined pre-launch, telemetry in place, a Day-30 review on the calendar. The agent watches dashboards, surfaces drift, and routes evidence into the next sprint.

The loop doesn't reset — it picks up where the last round left off. Conversations stay human; everything around them compounds.

// the.loop.detail DESIGN · SHIP · LEARN

Design

// spec → wireframes → clickable prototype

requirements.md
design.md
prototype.html

Ship

// pair-coded with Kiro · Cline · Q Developer

components
design-system
live-url

Learn

// launched, instrumented, evidence in

telemetry
user-sessions
evidence-backlog
The same loop, rendered as artifacts — what every phase actually produces.

Months. Weeks. Hours.

The compression isn't linear — it's compounding, because every agent makes the next agent faster.

12–16 wks
Legacy idea-to-handoff cycle.
1–2 wks
With agents in flight: synthesis automated, prototype live by mid-week.
hours
Mature cycle: idea → testable prototype in a single working session.

What the agent doesn't do.


The current toolkit.

The tools change every two years; the principles don't. Underneath every project: a spec layer, a design system, an audit loop, and a human in the chair.

Agents // reasoning + generation

Claude
Amazon Nova
ChatGPT

Orchestration // spec-driven build

Kiro
VS Code
Cursor
Cline

Craft // design + prototype

Figma
FigJam
Figma MCP

Loop // research + ship

UserTesting
Notion
Asana
Git · GitHub
AWS

One closing rule.

Process isn't a script — it's a posture. Same insistence on craft. Much more leverage. The lifecycle that used to live across a team now lives in one practice, multiplied by the agents alongside me.

Want to see the loop applied? The selected work →