Designing a scalable enterprise platform for real estate, construction, and design lifecycle management
Impact at a Glance
Impact at a glance
• Secured senior leadership buy-in and additional funding through prototype-driven reviews
• Enabled planned rollout across North America and Europe
• Reduced delivery timelines by multiple weeks by resolving ambiguity before development
• Established a scalable foundation for future AI-driven design optimization
Context
Amazon’s real estate and construction teams make long-horizon, high-cost decisions using fragmented standards, templates, and optimization tools. These workflows evolved rapidly during periods of high growth, but became rigid and difficult to scale globally.
As a result, teams struggled to reuse design knowledge, evaluate alternatives early, or adapt designs to site-specific constraints across regions.
Problem
Existing workflows required teams to piece together information from disconnected tools and documents. Constraints were often discovered late in the process, leading to rework, delays, and underutilized assets.
Users spent more time validating information than making confident decisions, and only a small percentage of available sites fit existing design templates.
Goals
The goal was to design a scalable system that could support proactive, data-informed decision-making across regions and programs.
From a business perspective, this meant increasing viable site matches, reducing time from site identification to launch, and enabling adaptive reuse and sustainability targets.
From a UX perspective, success meant reducing decision friction, supporting expert users without overwhelming them, and building trust in system-generated recommendations.
Role
I served as the Senior UX Designer and UX lead for Project MOSAIC.
I owned the experience end-to-end, from discovery and stakeholder research through high-fidelity prototyping and validation. I partnered closely with product, engineering, and data science, aligned stakeholders across North America and Europe, and presented the work directly to senior leadership to drive funding and roadmap decisions.
Process
Given the complexity and ambiguity of the problem space, I relied heavily on visual artifacts to drive alignment.
Early low-fidelity flows were used to clarify scope, define system architecture, and identify required data attributes before engineering ramp-up. These flows helped stakeholders move from abstract requirements to concrete decisions.
High-fidelity designs and an interactive prototype became the primary artifacts for leadership reviews, roadmap alignment, and cross-region coordination.
Strategy
The experience strategy focused on reducing cognitive load while preserving flexibility.
Key principles included establishing a single source of truth, introducing progressive complexity, and making system logic explainable and traceable. Users could start with simple actions and progressively access deeper system logic only when needed.
Design
MOSAIC unified five major capabilities into one cohesive platform:
• Design Templates
• Kit of Parts (standards and components)
• Build-a-Kit (customized guidance packages)
• Optimize Design (location-based recommendations)
• Design Generator (data-driven design variants)
The interface was designed to support browsing standards, assembling kits, optimizing designs based on context, and comparing alternatives across cost, throughput, and sustainability metrics.
Validation
Validation was critical due to the high-stakes nature of the decisions this platform supports.
I designed a usability testing and Customer Effort Score framework covering core workflows. We tested realistic scenarios using the interactive prototype and measured task success, time-on-task, and perceived effort.
This created a shared, quantitative definition of “good UX” across product and engineering.
Outcomes
The prototype directly enabled leadership alignment on product direction, secured additional funding, and unlocked expansion to both North America and Europe.
Beyond delivery speed, the biggest impact was shifting teams from document-driven workflows to confident, data-informed decision-making earlier in the process.
Reflection
This project represents how I approach complex systems: grounding design in real decision-making, balancing flexibility with governance, and using UX as a strategic lever not just an execution function.
If continued post-launch, my focus would be on deepening AI explainability, expanding UX metrics, and strengthening design system governance.