Dual-degree student at UVA studying Computer Engineering and Mathematics — building at the intersection of ML research, systems engineering, and real-world impact.
I'm a Computer Engineering and Mathematics student at the University of Virginia (GPA 3.76, Dean's List), with a minor in Philosophy and a year abroad at the University of Hong Kong. I'm drawn to work that sits at the intersection of technical rigor and real-world impact.
I've done ML research predicting material band gap properties using neural networks, built satellite testing infrastructure at ST Engineering iDirect, and shipped a first-place laser weed-removal system in an international engineering competition.
Incoming Software Engineering Intern at Appian this summer — always looking for the next hard problem worth solving with elegance and intention.
A semantic segmentation model deployed on an AMD Kria KV260 FPGA that labels every pixel of a reef frame as live, bleached, degraded or other — cheap enough in power to belong on an autonomous underwater vehicle. I own the deployment pipeline: quantizing the model to INT8, compiling it to a DPU graph, and getting it running on the board. As of September 2026 it holds 12.99 FPS at 76.99 ms per frame, with under 1% latency spread across 200 frames. Verified against an Unreal Engine digital twin of a real Indonesian reef.
Safety firmware for the tractive battery on Virginia Motorsports' Formula Student car. It watches all 144 cells and 48 thermistors across a galvanic isolation boundary over isoSPI, checks the pack against the competition's five fault conditions every 16 ms, and latches the shutdown circuit open when anything reads out of bounds — with no path in firmware that clears it. I lead the BMS firmware: the FreeRTOS task architecture, the cell-monitor driver, data acquisition, fault logic, charging handshake and CAN.
Designed and built an autonomous weed removal system under the theme "Smart and Sustainable Engineering." Used YOLOv8 for real-time object detection to identify weeds, then drove an Arduino-controlled servo motor system to target and eliminate them with a laser — calibrated for precise coordinate mapping from camera frame to physical space. Presented to a panel of industry experts alongside a multicultural team of 4 from Thailand and Hong Kong.
Comparative experiments evaluating training dynamics of gradient descent and diffusion-based methods across MLPs and 1D CNNs on a synthetic S-shaped dataset. Investigated model depth, layer width, and activation functions systematically.
Led all software initiatives for a mobile app that detects parasite eggs using a Roboflow object detection model. Built a RESTful API with Python Flask, communicated directly with the client to gather requirements, and delivered the final product demo.
Built at the UVA Women in CS Hackathon — a mobile app that lets users upload photos from their camera roll, detects foreign language text via OCR, and translates it to English in real time.
Whether you have a project in mind, a role to fill, or just want to connect — I'm always open to a conversation.