Emmaline Mai

Click through to see other work!

My Role:
As a research assistant I made various contributions to our mobile app, Peerdea and worked with the local Pittsburgh community to co-design it. My contributions included

• Enhancements (e.g., accessibility implementation, UI improvements)
• Maintenance and repair (e.g. bug fixes, code maintenance, unit testing)
• Oversaw deployments (e.g., builds for iOS and Android)
• Peerdea interview + workshop facilitation @ Prototype PGH
• Low and high fidelity prototypes for new features to help entrepreneurs progress towards personal/business goals through regular self-reflection
• Exploring large language models as a solution for quick feedback

Project Context:
Peerdea is a mobile app for IOS and Android that allows entrepreneurs to share designs and feedback on their creative work with each other. Peerdea builds on this by also supporting entrepreneurs in reaching their individual goals through regular reflection. In collaboration with the feminist makerspace Prototype PGH, over 3 years, 31 entrepreneurs have used Peerdea through an incubator.

See the Figma prototype here or read the paper here!

My Role:
I ran experiments comparing robot locomotion with/without various types of directional compliance locomotion by collecting, graphing, and analyzing data on feedback from sensors on the robot. I noticed correlations between torque sensor readings and when the robot hit obstacles, and developed a mass-spring-damper controls system that regulates the snake robot's amplitude based on those torque readings.

Project Context:
This snake robot requires pushing off of obstacles in order to propel itself forward, but complying with obstacles when they get in the way. Directional compliance determines when the robot should be pliable and when it should push against obstacles.

I created this dashboard as a personal project to support myself through the process of practicing cardistry. I had extremely minimal knowledge of front-end/back-end technologies prior to this, and also wanted to explore making my own web app. It recommends and displays YouTube tutorials of new moves for me to learn using the Google Data API, and when I'm learning moves I can add, edit, or remove them from a Microsoft Azure database. Finally, the dashboard suggests combinations of moves for performing using features of the moves in the database and a Markov chain.

See the Github repo here!

My Role:
I worked on the Android app, and helped build the screens that allowed available tutors and tutees to select who to connect with, and allowed tutees to rate tutors after just finishing a call. This was my first exposure to back-end work, and I introduced myself to SQL through working with this project. However, I wasn't satisfied with my contributions to our final product and this led to me building my own web app later.

Project Context:
This is a group hackathon project that lets people create accounts as tutors or tutees, then search for instant tutoring over video call through either the Android app or the website.

See the Devpost here and the GitHub repo here!

This project is still in progress. Currently, it makes a lot of API calls to various public APIs (Wikipedia for the main image, PaleoDB for lots of facts, the random dino from https://dinosaur-facts-api.shultzlab.com/). The immediate largest next step for me is to perform webscraping to build my own database of dinosaurs, and also make sure errors are handled correctly (eg. with a dinosaur not existing on Wikipedia).

Project Context:
After visiting the Carnegie Natural History Museum I wished I knew more random dinosaur facts off the top of my head, so I hoped this extension would expose me to more random dinosaurs.

See the GitHub repo here!