Paper++: Redesigning Academia and Paper-based Code Examinations
For Paper++, I utilized the Google Cloud Vision API to transform handwritten code from images into compilable and executable programs,
addressing a niche need in academic and professional settings.
This system supports user programs with static dependencies, offering a seamless transition from paper to a digital platform of a remote code execution system.
The project aims to streamline the grading process for paper-based code assessments and interviews.
I developed a Java backend to manage the automatic, thread-synced compilation and execution
of these programs, integrating the outputs into a React-based UI.
This project introduced me to leetcode's architecture with how I could improve Paper++ to become scalable. I hope to revisit this project and implement
a stateless architecture, using Docker compose for Boss-Worker nodes and load balancing.
Our efforts were recognized at MadHacks 2023,
where "Paper++" was ranked in the top 5 out of 58 projects,
validating the project's potential to modernize the evaluation of coding skills.