Lucy Mou

Hello, nice to meet you!

I'm Lucy, a Bay Area native with 8 years of design experience across fintech and B2B software.

I love untangling complexity and turning it into intuitive tools. I want to reduce the time and effort it takes to do complicated and/or tedious tasks, and create tools that let people do things they never imagined they could do. I want to give people confidence to act and time to focus on what matters to them.

When I'm not designing or prompting Claude, you can find me digging balls on the volleyball court, baking sweet treats, and hunting for the best boba wherever I go.

My design principles

Conviction with curiosity

I ground my point of view in research, but I hold it loosely. Strong team decisions matter more than being right.

Progress over perfection

I'd rather ship and learn than wait for the perfect solution. Set direction, make calls, iterate.

Open to all ideas

Good ideas come from anywhere. I evaluate every suggestion by the same criteria: user impact, feasibility, and business goals.

How I use AI tools

I use AI throughout my process, from clarifying rough ideas to building working prototypes.

For example, I built a calorie tracking app since existing ones didn't fit my needs. I used Claude as a thinking partner to clarify my idea and write a PRD. Then I used it to generate user flows and wireframe prototypes. I iterated with Claude, asking it to critique my designs and think through interactions with me. I moved into Figma for higher fidelity, then used Claude Code to build the app. I don't know how to code, so this let me make an idea a reality on my own.

I'll bring this approach to my next role, using AI to think through ideas and build clickable prototypes quickly, so I can communicate clearly and move faster with my team.

How I approach designing AI tools

Building user trust is one of the hardest problems in designing AI products so I use a framework to be transparent and let users keep control. I consider how confident the AI is, and how reversible the decision is.

  • If the AI is confident and the decision is easy to reverse, let AI act. QuickBooks auto-categorizes transactions, even if it's wrong, the user can easily change it.
  • If the decision is hard to undo or affects someone else, the user stays in control. QuickBooks AI agents will draft an invoices, but the user has to send it. Sending an incorrect invoice could damage their relationship with customers.
  • If the AI isn't confident, it should say so. AI can still share its best guess, but the user makes the final decision.
Get in touch

Let's work together.

I'm currently open to new opportunities. Whether you have a role to discuss or just want to connect, I'd love to hear from you!

Email me LinkedIn