cv

Software Engineer

Microsoft

• Developed internal API endpoints for Microsoft Defender XDR customer onboarding, offboarding, and metadata retrieval using Azure Functions
• Utilized React with TypeScript to develop a responsive frontend for the Microsoft Defender XDR onboarding wizard and settings page
• Rewrote legacy Kockout code in React for Microsoft Sentinel settings and pricing pages
• Contributed to the upkeep of services by implementing monitors, writing unit tests, and contributing to internal bug bashes

Bachelor of Science

University of Wisconsin - Madison

Computer Science

• CS 537: Introduction to Operating Systems
• CS 576: Introduction to Bioinformatics
• CS 540: Introduction to Artificial Intelligence
• Statistics 451: Introduction to Machine Learning and Statistical Pattern Classification

Machine learning principles applied to CT radiomics to predict mucinous pancreatic cysts

Abdominal Radiology

Current diagnostic and treatment modalities for pancreatic cysts (PCs) are invasive and are associated with patient morbidity. The purpose of this study is to develop and evaluate machine learning algorithms to delineate mucinous from non-mucinous PCs using non-invasive CT-based radiomics.

Pancreatic Cyst Classification

• Developed machine learning models to classify patients’ pancreatic tumors as mucinous/nonmucinous and malignant/benign to reduce unnecessary surgeries
• Created XGBoost and random forest models with oversampling and undersampling using Python and Scikit-learn
• Analyzed and processed 496 features for 103 pancreatic tumor patients from University Hospital’s dataset
• Acted as a lead of a 3-student development team to delegate tasks and maintain development schedule
• Utilized SHAP to analyze feature impact on model output

Renal AML Classification and Regression

• Guided two subteams of undergraduate students in developing machine learning models
• Developed and evaluated machine learning and Neural Network models to predict growth rate of renal AMLs and to classify renal AMLs as high or low growth

Renal Cell Carcinoma Neural Network

• Developed data transformation pipeline to clean data and make MRI scans usable for Neural Network
• Utilized transfer learning to train Convolutional Neural Network to classify Renal Cell Carcinoma as low or high grade