About Me
I am a front-end architect and UX/accessibility engineer who works in the translation layer between product intent, implementation reality, and AI-assisted delivery.
I help teams build complex systems that are not just functional, but usable, accessible, maintainable, and grounded in judgment.
Why translation matters
In complex product environments, the risk is rarely just whether a screen can be built. The harder problem is whether requirements, design intent, accessibility expectations, backend constraints, and implementation details still mean the same thing by the time they reach users.
From code to experience
I began my web career as a full-stack Java developer after earning a Computer Science degree. Early in my career, I worked on a technically useful application that struggled to gain adoption. That gap bothered me: the system worked, the functionality mattered, but people were not using it.
Trying to understand why led me to user experience. It also led me to Section 508 accessibility, at a time when formal training and structured learning paths were limited. I taught myself accessibility through available standards, guidance, and implementation practice, then deepened that direction through a Master’s degree in Human-Computer Interaction.
The work became translation
That experience changed the direction of my career. I moved deliberately from general full-stack development into front-end architecture, UX, and accessibility because I wanted to work where implementation decisions directly affect whether systems are understandable, usable, and responsible.
Over time, I became the person who could translate across disciplines: between backend constraints and front-end behavior, between visual design and usable interfaces, between accessibility requirements and practical QA validation, and between product intent and the code that actually reaches users.
AI-bilingual, before it had a catchy label
Today, that translation layer matters even more as teams adopt AI-assisted development. I think of this as being AI-bilingual: fluent enough in both domain context and AI-assisted workflows to use automation as a tool without surrendering judgment to it.
AI can accelerate scaffolding, refactoring, and exploration, but it still needs someone who can recognize when an output is technically plausible, structurally wrong, inaccessible, insecure, or simply not what the product needs.
Human, technical, and organizational languages
My work brings together computer science, HCI, front-end engineering, accessibility, and multilingual communication. I speak Cantonese, English, and Japanese, and I have always been comfortable learning new languages — human, technical, and organizational.
Ultimately, I help teams build complex systems that hold up beyond demos: interfaces that are usable, accessible, maintainable, and grounded in real product judgment.