AI-assisted
Live
UI Design
HokieHustle
Challenging developer's bias and turning a campus-wide problem into a local web solution.
TIMELINE
2 weeks
ROLES
UI Designer, Developer
TEAM
3 developers, 3 designers
SKILLS
UX interviewing, AI-coding
Overview
The Platform
The VTHacks Hacker App is for participants, judges, mentors, and organizers to handle registration, coordination and live event updates in one place.
The Problem
See progress as the user completes the application. Instant system feedback, compared to throwing an error upon a submit try.
Solution
Instant user-system feedback and progress indicators.
Added a humanistic and warm feel through content design (verbiage) and overall user interface.
Smarter and more intuitive question form types.
"We [designers] “touch” code not by manually typing every line, but by engaging in a fluid dance - moving from the design canvas to code and back again, using live prototypes to inform and mold our thinking in real time."
Jacquez, Katie. "Code Is a Design Material." Google Design
PROBLEM
The core problem surfaced early:
HEADER SECTION
s
Before
After
HERO
text
Before
After
INQUIRY
Affinity mapping
Raw interview data
88 work activity notes in
parent-child themes.
We also incorporated client-sourced data from a prior paper prototype study, which was what this project initially stemmed from.
Patients had no mental model for which portal held which records, leading to forgotten credentials and missed information at critical moments.
Responders needed critical details like allergies and current medications surfaced instantly, not buried behind a login.
Varying naming conventions across portals made navigation unpredictable, especially for vulnerable users.
Managing a dependent's care required role-appropriate access without compromising patient autonomy or privacy.
Persona strategy and setting the scene
SOLUTION
A Figma click-through prototype was built from low-fidelity sketches and a physical paper mockup. The prototype covered the app's core task flows.
Quantitative research methods we used
Scaling incrementally in difficulty to measure task completion accuracy, execution speed (seconds), or error rates
Subjective assessment to measure the perceived mental, physical, and temporal workload a user experiences while performing a task.
When collecting our subjective data during the empirical evaluation, we found some positive design choices and some areas of improvement.
Possible user conclusions we made
Novel apps reduce initial intuitiveness
The app progressively became easier to use, disproving the benchmark thesis
Lack of prior knowledge of the app limitation
Varied user mental models influenced expectations
Potential changes + Evaluation
COST IMPORTANCE TABLE
Buttons difficult to click
Difficult finding portal info
Lack of familiar patterns
Increase button padding/add icons
Group similar, recognizable information
Include onboarding experience
The project was successfully
in December 2025.