< Wrapper

Development Projects - Contract and Student Work

Development Projects - Contract and Student Work

Contract Projects

After graduation from my master program, it was important for me to keep up my skills and pay my students loans so I used my development knowledge to work on a few part time projects.

Military Wives Show Podcast iOS App

The Military Wives Podcast show requested an app for iOS where their listeners could locate information about how to contact them and to download and list the most recent podcasts. The design is pretty straight forward, built using Swift 4 in XCode. I ended up creating the icons and modifying some of the images using Pixelmator (cheap Photoshop). The end product will be on the app store in February 2018.

Psychology Clinic Assessment Tool

The Psychology Clinic at Washington State University hired me as a developer and statistician to build a website that would be used as an assessment tool. The site will be available online to the general public as well as onsite on a computer that is not connected to the internet. The general website is built using Javascript, HTML5, Node.js, and Python with an Apache web server.

It collects data about participants through a series of interactive surveys and then creates PDFs of the results for the participant and the therapist so that they can discus the results during sessions. It also posts the results to the server for the researcher, or to seperate files on the onsite computers.

Code is entirely available on GitHub.

Data Analysis for Small Software Company

I was hired by a small software company who could not afford a full-time data analyst for some business insights. I took 1 GB of data about users from their online demo, determined which features were the most important to tell if a user would buy a license, ran several analyses to determine what parts of what factors are most important for prediction, and created a presentation about what types of things might be of interest to their sales team and engineers. The trickiest parts of this process included, balancing a highly skewed data set and determining an analysis that could run given the size of the data and the limits of my laptop.

Student Coding Projects

During my time in the computer science I have started with under graduate courses and worked my way up to Machine Learning, Artificial Intelligence, and Reinforcement Learning Courses. iOS development is probably one of my favorites outside of those courses. Theory is also a lot of fun, even though it is the least favorite courses for more computer science students.

Reinforcement Learning Maze Game

I’ve made a lot of complicated projects now that I’ve started with the computer science department. One of my favorites is probably from my reinforcement learning course. The course only asked that we pick something that we wanted to use reinforcement learning on. During the class we talked a lot about one particular maze. It was used in every example of RL that we talked about. It was simple, but I often found it difficult to understand fully how the learners would process through the maze.

This program allowed students to go through each kind of learner to see how it processes and finds the best path. The plan is to make it even better and allow for more learners like SARSA and Q-learning so that the students for Fall 2016 can really see how it works.

For more information, see the paper that I wrote for the class.

Smart Watch for Nutrition Monitoring

I worked on a project with a team who were attempting to use a smart watch to track nutrition monitoring. I worked on the smart watch, on the code to parse the data that came from the smart watch, and the machine learning analysis. Check out the project paper here.

Marathon Training Application

I purchased an app on the app store from Hal Higdon. He has very good marathon training programs, but terrible iOS applications. It would take information about when your marathon was, but then when you went to post the training program to your calendar, it wouldn’t start from there and go backward. It would start from the current day, and go forward, and then end weeks before your marathon. So I built the app myself. The code for the project can be found on my GitHub.

Details about other projects or coding examples are available upon request, or via my github account.


Leah Zulas, PhD