$ ls ~/projects

Developed constraint-based optimization workflows using Google OR-Tools to automate vehicle characterization and sensitivity analysis for critical suspension parameters. Built a full-stack interface enabling engineers to run experiments and visualize results.

Lossy image compression framework using topological data analysis to preserve structurally meaningful features while discarding insignificant frequencies. Combines Fourier transforms with homology-based ranking of frequency components. Co-authored a prepublication arXiv paper comparing results against JPEG.

Momentum-based cluster rotation dashboard built with Streamlit to identify structural stock clusters and rotate into top-performing groups. Educational project — for demonstration only.

Wearable system to help visually impaired people navigate crosswalks independently. Lightweight ML pipeline using OpenCV and TensorFlow on an embedded Coral Dev Board for real-time edge inference. Developed under guidance of Ablr's visually impaired CEO, keeping user needs central throughout.

Autonomous robotics system for a 1:14 scale racecar at MIT Beaver Works, integrating LiDAR, IMU, and computer vision via ROS2 and OpenCV with multithreaded sensor fusion. Co-authored a published research paper on AI-driven pose estimation for emergency vehicle yielding, achieving 98% accuracy.