About me
I am a UXer who likes to (re)imagine how things should work – given real-world constraints – and test hypotheses with users. I go beyond mandates to maximize product impact, with a careful attention to detail.
I currently work at MathWorks, where I am responsible for designing and conducting user research about various aspects of MATLAB, a software platform for scientists and engineers. I partner with over 7 dev. teams and cross-functional partners to make MATLAB easy to use for non-programmers and flexible/powerful enough for developers. Some highlights:
- Spearheaded research activities to improve MATLAB’s startup experience: Interviewed and surveyed 25+ stakeholders and teams to reveal how they influenced MATLAB startup and identified multiple gaps. Conducted a mix of moderated and unmoderated usability activities to study a redesigned startup experience with 70+ users. Uncovered assumptions that didn’t hold true for 60% of users and evangelized changes across multiple teams. Establishing it as a repeatable process for other UXers to adopt for testing workflows.
- Performed a heuristic evaluation and convinced stakeholders against a redesign by illustrating how it would negatively affect target personas, and suggested viable alternatives.
- Designed new code creation workflows to alleviate multiple pain points and presented vision to senior stakeholders. Coordinated design and research efforts across disparate teams, pushing long-standing projects ahead.
Previously, I was a UX research intern at Google, where I redesigned how errors look in Flutter to improve readability and eventual task outcomes.
As a graduate student specializing in Human-Computer Interaction (HCI), I worked with Prof. Philip Guo in the Design Lab at UC San Diego, to create scalable user-centered systems to help people learn and do programming. I was also a teaching assistant for multiple HCI courses and conducted lectures for 200+ students. Here is an outline of the different areas of my work:
- Video interactions: reimagining interfaces to make programming tutorial screencasts easier to find, navigate and practice with, by extracting source code edits from the raw videos.
- Error messages: augmenting the coding-debugging workflow with learning aids when a programmer is stuck and needs help – among the ways is adapting ELIZA (a chatbot that mimics a psychotherapist) to aid self-reflection.
- Chatbots: evaluating the design of a conversational agent for augmenting learn-to-code UIs in place of live help.
- Permissions: designing a new model for Android apps that prevents misuse of permissions while reducing interruptions to the user.