UX at Amazon TFS: Scaling design thinking in a system that wasn’t built for it

Design in a Fragmented System

UX wasn’t just underutilized at TFS—it was structurally excluded. When I joined three years ago, design had no standard intake, no process visibility, and no seat at strategic planning tables. I was often the only UX designer on complex, high-impact systems with billion-dollar implications.

Bespoke Process as Default

At Amazon, problems don’t conform to templates. And that’s precisely what makes the work powerful.

When I first joined, I was frustrated by the absence of a consistent UX playbook. But now I understand:

There is no standardized UX process—because no two problems share the same architecture.

Each engagement becomes its own bespoke process. I calibrate methods to the maturity of the product, the nature of the user, and the tightness of the timeline.

Instead of asking “what’s the right process,” I ask:

“What is the bandwidth of this moment—and how do I design within it?”

I’ve adopted the same principle in life: show me the boundaries, and I’ll deliver clarity inside them.

Applied Judgment > Applied Template

This style of working requires trust in my judgment and communication. It means:

  • Preemptively scoping overflow
    • Predicting and accommodating upfront to avoid redesign, rework or system failure later. E.g. layout design that supports unlimited scaling (vertical stacking instead of tabs) that way the system doesn’t break or limit the user if reality exceeds the ideal. Building with foresight instead of just polish.
  • Designing for extensibility without overpromising.
  • Anchoring decisions in user goals, not stakeholder preferences
  • Knowing when to shift from “what’s ideal” to “what’s achievable.”

Constraints and Challenges as Patterns:

  • Zero-to-one product environments with no precedent
  • Inherited tech debt from PM-led or engineer-driven interfaces
  • Navigating stakeholders unfamiliar with design thinking
  • No design system, shared vocabulary, or consistent process

     

This is not just “follow standard best practices” but strategic design governance. And it’s become a core part of how I lead UX.

Instead of repeating those constraints in every project story, I’ve centralized them here—because what I want you to see is how I navigated them, not just that they existed.

Each case study that follows shows a different application of the same throughline:

  • Workshops & co-creation
  • UX Evangelism at Scale
  • Strategic Research and Planning
  • AI & ML in Design

 

Skip to the Case Studies:

My Problem Solving Process

01

Research & Understand

Excuse Deal say over contain performance from comparison new melancholy themselves.

02

Design / Solutioning

Excuse Deal say over contain performance from comparison new melancholy themselves.

03

Test & Iterate

Excuse Deal say over contain performance from comparison new melancholy themselves.

Image Not Found

Any questions find here.

New had happen unable uneasy. Drawings can followed improved out sociable not. Earnestly so do instantly pretended. See general few civilly amiable pleased account carried. Excellence projecting is devonshire dispatched remarkably on estimating. Side in so life past.

New had happen unable uneasy. Drawings can followed improved out sociable not. Earnestly so do instantly pretended. See general few civilly amiable pleased account carried. Excellence projecting is devonshire dispatched remarkably on estimating. Side in so life past.

New had happen unable uneasy. Drawings can followed improved out sociable not. Earnestly so do instantly pretended. See general few civilly amiable pleased account carried. Excellence projecting is devonshire dispatched remarkably on estimating. Side in so life past.

Hilary%20Law%20 %20Product%20and%20UX%20Designer%20and%20Problem%20Solver

Testimonials

Clients Testimonials

Hello👋i'm available for freelance work

For quick response: Chat now