Pixels to Players · Software Engineer
Computer Science and Statistics student at the University of Toronto with experience building full stack applications from the ground up. I have worked on projects that involve both backend logic and frontend interaction, focusing on creating systems that are functional, responsive, and user focused. I am comfortable working with databases, APIs, and application design, and I enjoy turning ideas into working products. I am especially interested in roles where I can collaborate with others, iterate on feedback, and deliver software that is actually used.
Pixels to Players
• Developed a client-driven playtesting analytics system through computer vision techniques to visualize player attention through real-time webcam-based gaze tracking and heatmap generation • Built synchronized webcam and screen recording systems using OpenCV and PyAutoGUI, capturing player reactions and interface interactions in real time • Implemented real-time eye tracking with MediaPipe facial landmark detection and custom PyTorch ML models (TinyMLP and regression mapping), improving previous accuracy by over 45% • Designed a scalable data pipeline processing 10,000+ gaze points per session for statistical analysis and heatmap generation using NumPy • Designed gaze-based metrics including fixation frequency, dwell time, and attention spread, delivering actionable insights stored in Firebase for game testing • Collaborated closely with development team and client utilizing Git workflows and Poetry environments to deliver an MVP within 6 wee
University of Toronto
• Full stack developer building visual tools to help instructors track and manage TA hiring workflows • Designed intuitive dashboards to summarize complex, multi-course hiring data in a single unified view • Added visual indicators to flag missing, incomplete, or inconsistent states for rapid review • Integrated frontend components with backend APIs to maintain state-aware visualizations across workflows • Tested edge cases to ensure visual outputs remained accurate across varied hiring scenarios
Squirl Signs
• Developing a machine learning model that translates ASL to spoken/written text in real time. • Involved in machine learning workflows including training models, analyzing performance, and hyperparameter tuning • Practical experience with Azure Cloud, including GPU based workloads for heavy tasks such as computing model training • Demonstrated ability to break down complex problems, implement robust solutions, and handle edge cases
Bachelor of Science · Computer Science
Majoring in Computer Science & Applied Statistics Minor: Mathematical Sciences.