People / Students / Class of 2024
Kexian (Kay) graduated with high honors from Stevens Institute of Technology in May 2019 with a major in Computer Science. Through research and internship experiences, Kexian developed a strong passion for extracting meaningful insights from extensive datasets. During her undergraduate research, Kexian actively contributed to several successful computational social science projects. One of the projects Kexian participated in from the beginning to successful publication was to examine 1,000 Wikipedia articles. She applied variable-length Markov Chains to a dataset of 858,101 article revisions. Using model-based clustering algorithms, she successfully extracted statistically significant routinized patterns. During her internship as a technology analyst at Credit Suisse, Kexian built applications for non-trading tasks like displaying real-time stock price percentage change charts and designed a Proof of Concept (POC) Fund Recommender System to help recommend financial products based on client transaction history and key product attributes. After graduating, Kexian joined as a full-time employee at Credit Suisse, where she gained valuable experience in working with extensive financial data. Collaborating closely with a market-making desk, she not only maintained and enhanced C# trading applications but also made significant contributions to trading outcomes by creating post-trade P&L trade analysis reports using Python and back-testing strategies to detect toxic order patterns. Additionally, she developed effective hedging methodologies using Python and C#, leveraging machine learning models and real-time/historical data for risk management. Following four years at Credit Suisse, Kexian pursued an OCR internship at Zhejiang lab in China to gain hands-on machine learning experience. There, she developed a bilingual OCR toolkit with high recognition accuracy, detecting and recognizing chapter information, formulas, tables, and figures from both Chinese and English Standards Documentation. The data was organized into JSON format to be fed into a large language model. Kexian is excited to be a part of the MLDS program at Northwestern and looks forward to expanding her modeling skillset and analytical mindset with a comprehensive curriculum and industry collaboration projects.
