See the whole system
Map the messy reality first — the data, models, tools, and the people who have to trust the output.
About Alex Wan
Six years making complex systems feel human.
I came up through industrial and interaction design, got hooked on systems thinking, and now shape how people work alongside AI.
How I work
This is where I come in.
Most AI products don’t fail on the model — they fail on the moment a person has to trust it. My process is built to find and fix that moment.
Find the real tension
Locate the exact moment where uncertainty breaks trust or blocks a decision. That is the problem worth designing.
Make it steerable
Design so people can see what the AI is doing, correct it, and stay in control instead of guessing.
Ship it and measure
Instrument the funnel, watch real behavior, and iterate until the numbers and the experience both move.
The short version
A few numbers behind the work.
Designing products
6 yrs Global brands shipped with
4 In AI product design
4 yrs International design awards
20+ Int’l & national juries
6 The path here
Each step moved toward harder systems.
I keep gravitating to the same problem in bigger and messier forms: how do you help people make confident decisions when the system is complex?
Microsoft Product Designer
I design conversational AI for Microsoft Fabric’s Data Agent — turning plain-English questions into trustworthy answers across SQL, DAX, and KQL. I built a reasoning-transparency framework spanning seven tool types and own the funnel from setup to answer. Shipped to GA in 2026, now serving thousands of organizations.
VMware Generative AI Design Lead I led VMware’s first Gen AI assistant from opportunity framing to launch — defining how it personalized complex answers, communicated AI-controlled navigation, and held a trustworthy enterprise personality, backed by a 44-component design system. Positive sentiment climbed from 0% to 75%; launched at VMware Explore with 200+ media mentions.
GEICO UI/UX Designer As the sole designer, I replaced four fragmented paths through damage inspection and repair scheduling with one eligibility-aware flow. Grounded in usability testing and hours of session replay, it lifted scheduling completion 10.2% and click-through 21% — proof that a calmer path pays off.
Penn Medicine UX / Product Designer In three weeks, with data scientists and a psychologist, I turned over a million COVID-era tweets into a public-health dashboard that made symptoms, sentiment, and geography legible. My first taste of making noisy data decision-ready — featured by Twitter Developer and The Philadelphia Inquirer.
Awards
Some of it came home with shiny things.
Three AI projects of mine earned a shelf’s worth of international design recognition — for making AI more human, inclusive, and trustworthy.
Judging
I judge design, mostly AI tracks.
I serve on design juries — often the AI and emerging-tech tracks — where I get to discover and champion aspiring designers doing bold, thoughtful work.
A personal loop
Games taught me to notice systems.
I’m a game nerd drawn to feedback loops, progressive disclosure, readable states, and the tiny interactions that help people understand what happens next. That curiosity follows me into every product I design.