Finance Graduate (B.S.) | Minor in Applied Computing
M.S. Computer Science Student | Machine Learning & Database Systems
Finance graduate (B.S.) with a minor in Applied Computing, currently pursuing an M.S. in Computer Science at the University of South Carolina. I have experience in full-stack web applications, Python, and computational mathematics. I developed strong problem solving skills and applied clean, efficient code while working on projects. Going forward, my academic focus is on machine learning and database systems, as I am passionate about building intelligent systems while exploring how mathematics and software engineering work.
A project that implements a pattern-based algorithm for factoring a specific class of semi-prime numbers. This program looks for connections between sequences: 1,5,9,13,17 and 5,45,117,221 along with others. This can expand further to work for gap-5 and gap-6 semi-prime numbers. My work of using algorithms that calculate the prime factors of numbers demonstrates my ability to conduct graduate-level research showcasing my capacity to identify and manipulate mathematical patterns, work with algorithms, and use creative problem-solving.
The sequence generator produces candidates for semi-prime factorization and validates output across very large number ranges. This work demonstrates my ability to conduct graduate-level research using computational mathematics while showcasing my capacity to identify and manipulate mathematical patterns, work with algorithms, use creative problem-solving, and work independently.
I developed a program that uses sentiment analysis to identify whether Amazon reviews are positive or negative. My program uses the Amazon Review Polarity Dataset (3.6 million reviews spanning 18 years), as well as machine learning and natural language processing. My model uses a Support Vector Machine (SVM) to separate two classes of data with a straight line (linear kernel) and TF-IDF vectorization to extract text. The system predicts sentiment polarity from the text with accuracy as was demonstrated on test data. My project applies NLP techniques to classify text for real-world application.
I used Discord bots to create an interactive horror experience, combining full- stack development and narrative game design. Players sign up for a ‘weather assistant’,Billy, through a retro terminal-style web interface. Billy gradually reveals sinister intentions through timed, branching conversation trees. A second bot, Kathy, intervenes to warn players. She creates a complex narrative with 12+ decision points and multiple endings. My project demonstrates advanced bot coordination with persistent state management across user sessions, asynchronous message timing for dramatic effect, and integration of web (Flask) and chat (Discord API) platforms. A fake social media profile (FaceLibrary) enhances engagement and the user is able to choose responses which develop dynamic conversations. JSON files allow multiple bots to communicate, and the aesthetics of the terminal along with the psychological horror elements serve to create an engaging player experience that showcases creative storytelling.
A second bot, Kathy, intervenes to warn players, creating a complex narrative with 12+ decision points and multiple endings. The project demonstrates advanced bot coordination, asynchronous message timing for dramatic effect, persistent state management across user sessions, and integration of web (Flask) and chat (Discord API) platforms.