Abstract
Introduction
This portfolio has two projects that show how technology works in the real world. It is not just about engineering or policy, but about how people, institutions, and systems all must work together. The first is a capstone project called G.L.A.S.S., smart glasses that help people with mobility limitations control smart devices using hand gestures. The second is an STS research paper called "Invisible Circuits, Visible Risks" about why FPGA security fails in high-frequency trading when the networks of engineers, vendors, regulators, and traders don't align. Although they are different in many aspects, they both make the same point: technology doesn't exist in isolation, and you must understand the whole ecosystem around it.
Capstone Project Summary: G.L.A.S.S.
G.L.A.S.S. (Glasses for Linked Automation of Smart Systems) is smart glasses that let people with mobility limitations control multiple devices through hand gestures. It uses two Raspberry Pi Zero 2W boards in a 3D-printed frame, a Raspberry Pi 5 as the main processor, Google's MobileNet V3 for detecting devices, and an OLED display to show feedback. You look at a device to select it and use hand gestures to control it (lights, TVs, and computer mice) wirelessly. The whole project costs around $183 to build. More importantly, it solves a real accessibility problem by giving people one simple way to control multiple devices instead of juggling remotes and dealing with physical barriers
STS Research Paper Summary
The paper looks at why FPGA security fails in high-frequency trading using Actor-Network Theory. The fundamental argument is that security doesn't just come from better technology or better laws, it depends on whether all the different actors (engineers, vendors, regulators, traders) are working together. Two cases are used to show this breaking down: the Aleynikov case from 2012, where Goldman Sachs' code got stolen and the courts were unsure how to handle FPGA configurations legally, and the Network-1 v. Optiver case from 2022, which shows the same problem continuously occurring because the system is fundamentally broken. Using ANT, I argue that these failures aren't just technical problems or legal problems or economic problems on their own, but they are about networks falling apart. The paper suggests that fixing FPGA security requires all these different actors to coordinate and pay attention to power imbalances, not just building better technology.
Concluding Reflection
Working on both projects at the same time taught me that you can't separate technical thinking from understanding how people and institutions work. Building G.L.A.S.S. meant really thinking about what users need and how to design something dignified and useful, not just something that technically works. The STS paper made me realize that technology failures are almost always about networks breaking down, such as people not communicating, incentives misaligned, and lack of transparency. Doing both projects together showed me that the best engineering and policy questions require holding both perspectives at once: not just "can we build this?" but "what has to align for this to actually work in the real world?" I'll carry that into whatever I work on next, whether it's assistive technology, defense, or anything else.
