Pose Ninja is an interactive, webcam-based game that uses real-time body tracking to encourage movement and fast reactions. Players tap randomly appearing targets with specific body parts while avoiding incoming bombs, testing their accuracy and speed. To ensure the game is widely accessible we designed the game to run efficiently on CPU-only hardware. The game supports both single-player and two-player modes.

This project involved programming a Franka Panda robotic arm to pick up blocks from a workspace and stack them on a target platform, with scoring based on block type (static or dynamic) and final stack height. We built a robust computer-vision–based pick-and-place pipeline optimized for static blocks, while also exploring both hard-coded and vision-driven methods for interacting with dynamic blocks.

For this project we built Bingus Search, a multi-threaded HTTP web server and search engine in C++. The goal was to create a system that could efficiently index files, handle multiple simultaneous users, and provide both web-based and command-line interfaces for interacting with files on the server.

This is a Stirling Engine modeled as a hamster running in a wheel. The engine was designed in SOLIDWORKS and most parts were machined at UPenn's Precision Machining Laboratory using, Mill, Lathe, and CNC Operations.

In this project, we analyzed over 400,000 NYC restaurant inspections to predict inspection grades using historical data and engineered features, while identifying which factors most strongly influence inspection outcomes. By highlighting feature importance, we provide restaurant owners with actionable insights to proactively address issues that could result in lower grades. Ultimately, our framework seeks to improve public health by empowering diners to make safer, data-informed choices based on historical restaurant performance.

We built a fully self-contained Connect 4 system where a human can play against an AI without any laptop or cloud. Using a Seeed Studio XIAO ESP32-S3, an onboard camera, and a lightweight CNN, the system reads the board and uses a Minimax algorithm with Alpha–Beta pruning to choose the best move, which is shown on an LED board. This demonstrates full AI gameplay running on a single microcontroller.