Deven Wells
Engineering Design & Engineering Management
Aussie Exchange is a Java-Swing web application that pairs older adults in Japan with friendly Australian visitors for casual conversation, cultural sharing, or guided experiences. The goal: ease loneliness in Japan’s aging population while giving tourists a richer, people-first way to explore the country.
DEMO:
My Roles
Java-Swing Developer
- Implemented language-switching, authentication, and accessibility tweaks
- Refactored UI components after each round of user testing
Project Manager
- Set weekly milestones and kept the team of four on schedule
- Facilitated stand-ups & retros to maintain a healthy team dynamic
Documentation Lead
- Authored the localization summary, usability-test reports & final slide deck
- Ensured all deliverables met course and ISLO standards
The Problem
-
Social Isolation: Japan has the world’s fastest-aging population; single-person households and reported loneliness are rising.
-
Tourism Opportunity: Australian travel to Japan has grown 5× since 2010. We saw potential to connect these two trends for mutual benefit.
The Solution
-
Profile Browsing – Japanese users scroll a card-style list of vetted Aussies.
-
Secure Matching – Requests are encrypted; both sides approve before any contact.
-
Language Support – One-click language swap (日本語 / English); all labels auto-localize.
-
Accessibility First – High-contrast palette, scalable fonts, keyboard-navigable layout.
Cultural Localization Approach
We wanted to build more than just a translated app, our goal was to make Aussie Exchange feel natural and intuitive for Japanese users.
Research-First Process
Started with academic and online research to understand social issues in Japan (like loneliness among older adults) and cultural norms around design, privacy, and interaction.
Interviews with Cultural Experts
We talked with Japanese users and cultural experts to ensure our concepts addressed a real need, not just what sounded good to us.
Usability Testing:
Round 1
Our first prototype (Figma) focused on basic flow and structure. Feedback helped us fix spacing, labeling, and other quality-of-life issues.
Results
Usability Testing:
Round 2
The second prototype was built in Java Swing with full functionality. We tested it with four Japanese users (either native or fluent). Their insights changed everything.
JP Font Matters
Default Java fonts appeared Chinese. We switched to MS Mincho, a Japanese-friendly font, for all localized screens based on direct user feedback.
Replacing Commercial Imagery
Our original login screen used a Mt. Fuji photo that felt too touristy. We replaced it with friendly, human-centered visuals to match the app’s purpose.
Improved Interaction
Users tried clicking on flag icons to change languages, but only the small button below worked. We made the whole flag clickable for easier use.
Sorting Differences
Unlike U.S. apps like Amazon, Japanese users use fewer sorting filters. We simplified our browsing experience based on that preference.
Tech Stack
Frontend & Backend:
Java 17 + Swing
Authentication System:
Custom-built using CSV files and secure password hashing
Version Control:
GitHub Classroom
Create Your First Project
Start adding your projects to your portfolio. Click on "Manage Projects" to get started
Automated Turret
Project type
Mechatronics Project
Date
May 2024
Location
Terre Haute
For our freshman design capstone project, we were required to create a project that uses code with sensors, motors, and actuators. With the idea of using computer vision in a project like this, we chose to go with an automated turret. In this project, we had many stages of the process that were quite difficult but rewarding to overcome. I was mainly tasked with the coding, sensor, and motor controlling. My work greatly helped me better understand the basics of computer vision and how to use a Pico to wirelessly control a mechatronic system. At the end of this project, we were given the opportunity to showcase our project at the annual Rose Show. Even though we did not receive any awards, it was a great opportunity to learn about computer vision, microcontrollers, and refine my teamwork skills. Reflecting on the project, I would change from using the HuskyLens to using a regular camera with CV code to track objects instead. This would be much more complex than using the HuskyLens; however, it would allow for better tracking of objects at a distance with more precise movements. Other than this hardware change, I would say the project went great technically and personally with the team. In other images in the portfolio entry, you can read more about the process and the projects.
