As my thesis project I am working on music information retrieval (MIR), evaluating the benefits and drawbacks of various designs including combinations of CNN, RNN, GRU, dilated layers, residual blocks, and more!

This project has let me examine both my passion of music and AI, my end goal being to provide a useful tool for artists.

Working as part of a large multi-disciplinary team led by P. Fu, A. Wilkinson, V. Mago, R. Schiff, and A. Fisher, I contributed to the 2024 Point-in-Time count in Thunder Bay.

I helped with the technical side of the data-entry process, and it was rewarding to see how our collective work could provide more accurate data to help drive social policy and community change.

In collaboration with Dr. Diyab, Ben Fedoruk, and Ahmad Diyab, I explored how Large Language Models (LLMs) can be safely integrated into the classroom. We developed "AI Assess," a system that uses prompt engineering and the ChatGPT API to automate grading and identify student knowledge gaps, with the goal of reducing instructor workload.

It was a great experience working with the team, discovering the strengths and limitations of ChatGPT in educational assessment.

I worked alongside Ben Fedoruk, Harrison Nelson, and Kai Fucile Ladouceur to research a "democratic" alternative to traditional social media moderation. Our team looked to see if there was a link between sentiment and how misinformation spreads across platforms like Reddit and 4chan, proposing an algorithm that uses a "jury" of anonymous users to flag content.

This project let us explore the intersection of social trust and data science, and I really enjoyed contributing to our team’s effort to find technical solutions for online safety.