Research Assistant in Computer Vision
Research Interests
My research focuses on robust 3D perception for autonomous systems. I developed "Difficulty Painting," a multi-modal fusion framework that projects learned detection difficulty scores onto LiDAR point clouds. This method functions as a 3D attention mechanism for hard-to-detect objects, achieving state-of-the-art performance on the KITTI benchmark. Complementing this applied work, I explore Scientific Machine Learning utilizing nonlinear dynamics to improve neural architecture efficiency.
Education
- Honours BSc, Computer Science & Mathematics, University of Toronto, September 2021 – June 2025
Relevant Coursework
Machine Learning, Deep Learning, Neural Networks, Reinforcement Learning, Linear Algebra, Probability, Statistics, Algorithms, Data Structures, Operating Systems, Database Systems, Optimization, Numerical Methods
Skills
- Languages: Python, Java, JavaScript/TypeScript, SQL, C++, C#, Go, R
- ML&Data: PyTorch, TensorFlow, Pandas, NumPy, Scikit-learn, PySpark, Neural Networks, RL, Scientific ML,
- Image Processing, Pattern Recognition
- Cloud/DevOps: AWS, Docker, Kubernetes, Terraform, CI/CD, Azure
- Other: Algorithms, Data Structures, REST, Agile/Scrum, English/Mandarin