Uno (AWS Fintech)
I led design for UNO, AWS Fintech's flagship AI intelligence platform — unifying 27 disconnected tools into a single platform serving 30,000+ users and managing $2.75B+ in spend.
UNO
The Story
Amazon's finance organization had a trust problem. Analysts navigated 27 disconnected tools for routine analyses, each taking 2–4 hours. Different tools produced different numbers for the same customer. Senior leadership was making billion-dollar decisions on data nobody fully trusted.
I was brought in as design lead to fix this — not by building tool #28, but by replacing all 27.
Understanding the real problem
I ran 15 contextual inquiries with finance managers and discovered something the briefs missed: the core issue wasn't feature gaps in individual tools. It was context-switching. Analysts spent more time navigating between systems and reconciling data than actually analyzing it. The insight shifted our entire approach from "better tools" to "one intelligent platform."
Designing the intelligence layer
UNO consolidated everything into a single hub with 99% data accuracy. But the breakthrough was "Search by Jarvis" — a natural language interface (Text-to-SQL) that let analysts ask questions in plain English instead of hunting through dashboards. A 4-hour workflow became 45 minutes.
Scaling trust
With one source of truth, leadership could finally make decisions confidently. UNO became the daily insight tool for Amazon and AWS CFOs — featured in the Wall Street Journal. The platform's success drove 5x growth in R&D investment for AWS Fintech.
But UNO was still reactive — analysts had to ask it questions. The next chapter was making it proactive. (See: GDSP Workflow)
The Impact
