Hello, I'm
Full-Stack Developer & CS Researcher at Bucknell University
I'm a Computer Science & Engineering student at Bucknell University with a passion for applying AI, optimization, and formal methods to hard real-world problems.
My research spans adversarial ML, 3D mesh compression, EEG signal analysis, and formal program logic. Outside the lab, I've tutored 100+ students in CS, Calculus, and Physics.
Studying Guarded Kleene Algebra with Tests (GKAT), a formal system for modeling program logic and control flow with applications in autonomous systems and AI agents.
Investigating hierarchical multigrid methods for 3D mesh compression with adaptive basis optimization. Implemented prototype in PyTorch + C++ integrating CGAL for geometric processing.
Developed MILP models for adversarial neural network verification. Led a team of 3 to build reproducible deep learning & optimization workflows. Deployed on HPC cluster, reducing experiment time by 30%.
Tutored 100+ students across CS, Calculus, and Physics. Led weekly study groups for CSCI203/204, Calculus I–III, and Physics 201/202. Mentored new tutors on pedagogy.
Analyzed deep learning algorithms on EEG brain-signal data to detect and mitigate gender bias. Preprocessed datasets with Pandas/NumPy and reproduced PyTorch models from prior research.
MILP-based framework for formally verifying neural network robustness under adversarial perturbations. Integrated Gurobi heuristics to significantly reduce solve time on benchmark instances.
View on GitHub →An AI-driven website generator using multi-agent coordination and static code analysis. FastAPI backend deployed on AWS (S3, EC2, CloudFront) for scalable build workflows.
View on GitHub →Hierarchical multigrid methods for 3D mesh compression with adaptive basis optimization. Evaluation pipeline using Chamfer distance for reconstruction quality assessment.
View on GitHub →I'm always open to research collaborations, internship opportunities, and interesting projects.