Ying Zhou

  5+ years of experience in machine learning/statistical modeling[YZ1] , data analysis and developing production code.

Techlent, Data Scientist Fellow                                                                                                          Dublin, CA, 2022 - Present    

• Built a forecasting model to identify fraudulent healthcare providers when submitting claims for reimbursement.
• Cleaned and preprocessed ~600,000 medical claim records; engineered categorical features out of original features containing over 12,000 diagnosis/procedure codes. Used up/down sampling to handle the binary classification imbalance problem. 
• Evaluated logistic regression, random forest, XGBoost model via AUC score; optimized a random forest classification model to detect potential fraud (93% precision),  identified top important features correlated to fraud activities and saved $50M of claim reimbursement amount based on the dataset.
• Implemented end-to-end model pipeline and deployed final model as an API on Google Cloud Platform.

 Workforce Software, Software Engineer                                                                                    Livonia, MI, 2017 - Present                                                                               

• Collaborated with product and design teams, developed and deployed a micro service as part of the SaaS workforce management platform, using Java, SpringBoot, Kafka, MySQL. The platform processed, aggregated, and displayed employee work and attendance data for over 200 customers with ~2 million employees.
• Integrated MySQL into the service to improve the data processing bottleneck. Increased the display limit from 200 employees to 5000 employees in the application after implementation.
• Developed new features and maintained an internal application using Java, Spring, SQLServer, and Python. Automated application upgrade process using Jenkins.

 ClearServe LLC, Data Application Developer                                                                                            New York, NY, 2015

• Developed and maintained a web-based personal wealth management application. Integrated financial data ETL, aggregation, and visualization. Generated custom report using Java, MySQL. 

 Cornell University, Research Assistant/Data Analyst                                                                       Ithaca, NY, 2013 -2014

• Built a K-means clustering model to discover digital calendar scheduling patterns from ~16,000 Google Calendar scheduling records; implemented a pipeline for data ETL and engineered time features; separated the user scheduling patterns into 3 major clusters and provided suggestions to app design team.

 University of Massachusetts Amherst, Graduate Research Assistant                                 Amherst, MA[YZ1] , 2005 - 2012

• To investigate the function of a novel gene, used statistical methods (A/B Testing, chi-square test, ANOVA) to design experiments, collect, analyze, and visualize data; Uncovered the molecular mechanisms of how this gene induces cell death, provided new strategies to the treatment of cancer and neurodegenerative disease. 
• Conducted bioinformatics analysis of this gene family, performed hierarchical clustering analysis on the DNA sequences and visualized the phylogenetic tree.  Developed functional studies based on the evolutionary study.