Developed and demonstrated prototypes of plagiarism detection tools that analyze 2-second code snapshots to identify potential plagiarism in diverse scenarios, including exercises with extensive starter code and creative tasks. Applied a logistic regression model, detecting about 2% of submissions as plagiarism with 0% false negatives in the test dataset.

Conducted experiments with approximately 60 volunteers to collect data and applied various machine learning models to predict instances of plagiarism.

Led the development of the model, and handled all UI/UX design aspects of the product, ensuring a seamless user experience.