Abstract
This portfolio has two projects I worked on during my spring 2026 semester at the University of Virginia. The first is StrideSense, a wearable device my team built to help patients recovering from knee injuries, athletes, and physical therapists get real-time feedback on knee loading without needing expensive lab equipment. The second is my STS research paper, which looks at why residential solar energy in the United States still has not reached the communities that need it most, even though the technology has gotten much cheaper over the years. When I first started both projects, I did not think they had much to do with each other. But the longer I worked on them side by side, the more I kept running into the same question: why does technology so often end up helping people who already have the most, instead of the people who actually need it? That is the thread that connects everything in this portfolio.
Capstone Project Summary:
StrideSense is a low-cost wearable system that monitors knee joint loading and detects abnormal walking patterns in real time. The problem we were solving is simple: if someone recovering from a knee injury wants to know whether they are moving correctly during rehab, the only reliable way to find out right now is inside a clinical lab that most people will never have access to. Our system attaches two sensors to the leg that collect acceleration, orientation, and angular velocity data. A microcontroller board filters and processes that data, saves it to a memory module, and sends it to a computer over USB, where a machine learning model estimates the actual force on the knee. The user then sees the results on a mobile interface that tells them whether their loading looks healthy and when they should rest. My role focused on data processing research and staying in contact with our clients, including physical therapists and patients, to make sure what we were building would work for real people. The hardest parts were calibrating two sensors accurately, making the ML model reliable enough to be useful, and filtering out noise without losing important data.
STS Research Paper Summary:
My STS research paper asks what social barriers keep people from adopting residential solar energy in the United States. The topic felt personal to me. Growing up in Afghanistan, I always assumed solar was out of reach for most families because it was a developing country. Then I moved to the United States and noticed the same unequal pattern. As a renter myself, I ran into one of those barriers directly, and that experience is what pushed me to research what was actually driving the gap. The paper uses the Social Construction of Technology framework, which argues that whether a technology gets adopted depends less on how well it works and more on the social conditions around it. Through reviewing academic research and policy reports, I found four consistent barriers: financial inequality that makes upfront costs and tax credits inaccessible to lower-income households; the split incentive problem that locks renters out because they do not control the roof they live under; political and geographic differences that determine how much government support a household can actually access; and institutional policy barriers like complicated permitting processes and incentive programs designed around wealthy homeowners rather than the people who need the most help.
Working on both projects at the same time taught me something I would not have gotten from either one alone. When I was deep in data processing for StrideSense, most of my focus was on whether things worked. But my STS research kept bringing me back to a harder question: even if it works, who is actually going to be able to use it? That question came from months of reading about how solar technology keeps ending up with people who already have financial stability, no matter how cheap the panels get. It made me think about StrideSense differently. We wanted it to be affordable, but affordable and accessible are not the same thing. A device that assumes a certain type of patient or clinical setting can still leave people out even if the hardware is cheap. On the other side, working on the technical project made my STS arguments feel more concrete and real. After struggling with sensor calibration and noisy data, I understood in a very direct way why the SCOT framework makes sense. The technology is almost never the hard part. The hard part is everything built around it. I am finishing this semester thinking about engineering differently than when I started, and I think that is exactly the point.
