About

I am a PhD student in computer science at the University of Alabama, in Chris Crawford’s Human-Technology Interaction Lab. My work sits between two things that look separate and are not: radar sensing, and how people learn machine learning.

The sensing side is micro-Doppler. A 60 GHz radar sensor returns a signal that, once you put it through a range FFT, clutter suppression, and a short-time Fourier transform, becomes an image of motion over time, a spectrogram in which a wave of the hand has a shape you can recognize. I care about what those systems can do as interfaces, and about the gap between how they behave in a lab and how they behave in a room full of people who did not read the manual.

That gap is where the education work comes from. Machine learning is usually taught on datasets that have already been cleaned, labeled, and balanced by someone else, which quietly removes every decision that matters. SensDS, the system I build, hands those decisions back: students collect their own radar data, train on it, and watch the model fail in ways that are legible. A novelty detector I added as an engineering feature, flagging captures that fall outside the training distribution, turned out to be the best teaching moment in the system, because it makes distribution shift something that happens to you rather than a term in a lecture.

Before Alabama I spent three years in Naja Mack’s Human-AI eXperience Lab at Morgan State, working on VR for concussion education, bias in conversational AI, and K-12 computing outreach, and coordinating the CodeBears summer camp across three summers. I taught Introduction to Computer Science I there as adjunct faculty in spring 2025. Most of what I believe about teaching came out of those rooms.

Where I am headed: sensing systems that are legible enough to learn from, and computing classrooms where the hard parts are visible instead of hidden.