Advised by Prof. Ming Lin, researching physics-based simulation of autonomous micro-mobility vehicle navigation through crowds.

Developing a cross-platform Flutter application for building and visualizing quantum circuits with Qolour.io. Architected the backend with Firebase, ensuring low-latency circuit encoding and enabling seamless execution of circuits on multiple quantum computing backends.

Led a 10-person team to engineer an automated weapon detection and rapid-notification app using React and Expo. Powered by multi-threaded CUDA-accelerated PyTorch inference, the system analyzes live security footage to provide real-time threat and suspect tracking. The product's innovation and technical execution earned recognition as a Finalist in the xFoundry school safety startup competition and a Semifinalist at the Texas A&M AI Pitch Competition.

Drove high-impact ML and system modernization initiatives across two projects. Optimized an internal Systems Engineering AI tool by integrating custom deep learning models and advanced LLM prompting via Semantic Kernel agents, which increased INCOSE rule detection F1 scores by 115% and reduced manual review times by 25%. Alongside building Streamlit evaluation dashboards and enhancing the MaaS framework for rapid deployment, I successfully executed eight zero-downtime upgrades to modernize the Enterprise Metrics Repository's data handling at scale.

Advised by Dr. Shabnam Jabeen, presented a poster on Hybrid Quantum-Classical Wasserstein Generative Adversarial Networks, demonstrating a 15% reduction in the Wasserstein distance using data augmentation on the MNIST dataset. Implemented a quantum support vector machine that achieved a 5% increase in accuracy for breast cancer classification compared to classical models.