In the summer of 2023, I had the opportunity to work as a research assistant for Dr. Juan Rojas. Our research focused on the usage of symmetries in the environment to improve learning rates for reinforcement learning. 

Over the course of the summer, I gained a lot of experience being self-driven to learn new concepts. I learned to use libraries like Numpy and PyTorch and gained experience working on designing and implement sophisticated algorithms.  

An outstanding issue in this field is that designers must specify how the environment is symmetrical rather than the agent learning it. I helped lead the team through designing and implementing an algorithm that allows agents to learn how their environment is symmetrical. 

In the summer of 2022, I released a game called (square) on Steam and itch.io. I collaborated with Brett Boyd who took care of the sound and music. 

The game is a mixture of puzzle and action. The player is armed with one projectile they can launch around the arena or call back to them. The projectile, however, is just as deadly to the player as it is to the enemies. 

The game has positive reviews on steam. For a more thorough look at the game, check out this video review of the game. 

During the Spring semester of 2023, I took on an extracurricular project with Dr. Bryan Crawley. 

GSTAL is a simple language that exists to be a target language in the compiler construction course.  Due to the low level of abstraction and the stack-based memory, manually writing code in GSTAL is a notoriously frustrating experience.  

The BC GSTAL Debugger aims to make it a less painful experience by allowing users to write, run, and debug their code with a GUI interface. Features include breakpoints, adjustable run speeds, and a visualization of memory and the current register values. 

It was written in Python with the tkinter library. The code is available here.

Using Unity, I made a program that simulates simple ant colonies. Ants will collect food, fight ants from other colonies, and starve without food. The user is able to interact with the simulation by placing food or stomping the ants. The program has been viewed over 4,500 times on itch.io.

Having learned much more about programming and design in the years since I made this, there are many things I would like to improve about it like more accurately simulation involving pheromone trails and general polish.

itch.io link