Work experience
- General Motors (July 2021 – Present) – Data Scientist
- Led the pilot development of a conversational chatbot using Langchain and GPT-4 to automate the generation and execution of SQL queries based on questions from the Portfolio Planning division to improve workforce efficiency by 45%
- Developed a survival analysis model to predict the scrappage rate for all vehicles on an annual basis with 73% accuracy in order to provide business intelligence into the demand for non-collision part repairs
- Led the development of a recommendation engine using statistical methods to boost the recommendation rate for vehicles that were previously constrained for dealerships
- Nationally-launched for all GM dealerships in USA
- Developed a random forest model with 63% accuracy to determine the amount of delay that a vehicle will experience during the delivery to the customer
- General Motors (Summer 2020) – Software Engineering Intern
- Developed UI pages using JavaScript and Vue.js and backend using Spring Boot for the HSM File Type Signing Project
- Used JPA to update and retrieve user and file data from a PostgreSQL Database
- Released the frontend and backend contributions into production for internal use
- Used market basket analysis to find correlations between car features and the time it takes that car to be sold
- Worked in an agile and scrum environment
- Social Dynamics and Wellbeing Lab (Spring 2019 – Present) – Research Assistant
- Used NLP techniques to train supervised machine learning models on glassdoor posts to predict company performance
- Processed and labeled social media posts to understand people’s expression of major life events through social media
- Scraped Twitter, Reddit, and Tumblr posts that contain CDC-provided keywords that correlate with suicide
- UPenn Operations, Information, and Decisions Department (Summer 2019) – Research Assistant
- Analyzed the impact of airballs on a basketball players performance using SportVU camera data from NBA games
- Developed software to identify airballs using player and ball coordinates from SportVU camera data
- Correlated changes in performance measures, such as shots, rebounds, passes, with airballs
- LKQ Corporation – IT Intern
- Computational Economics Research Lab (Summer 2017) – Research Assistant
Publications
- Development of a Machine Learning Model Using Multiple, Heterogeneous Data Sources to Estimate Weekly US Suicide Fatalities (Dec 2020)
- JAMA Network
- Crime Prediction and Optimal Police Allocation (May 2018)
- Young Scientist Journal
- Summer Research Symposium at Vanderbilt University: Honorable Mention
- Middle TN Science and Engineering Fair: 2nd Place Computer Science Award
- Development of basketball metrics using SportVU, an advanced optical recognition system (May 2016)
- Tennessee Junior Academy of Science Symposium: Runners Up Award
- Joint Retreat Research Symposium at Vanderbilt University
resume
Please feel free to view and download my resume for more details on my work and project experience!