Xiaoyi Wang
Principal Associate, Data Scientist @Capital One
Signup · Get unlimited contacts
WORK HISTORY
Principal Associate, Data Scientist @Capital One
New York, NY, US
Full-Stack Machine learning engineer with extensive experience in designing and deploying end-to-end models and pipelines. • Technical lead: Led 3 MLEs and 1 PM in developing a XGBoost machine learning model used to define optimal offer strategies for customers in delinquency. Responsible for design and implementation of a prototype from initial project ideation to a deployed production model.• Reinforcement Learning: Led the development of a Markov decision process reinforcement learning model that prescribes personalized contact strategies, generating $25M incremental NPV annually.• LLM Prototype: Created a contact email generation system using LLama to improve email open rates during an internal hackathon.• Machine Learning: Developed five proof-of-concept models using logistic regression and XGBoost for the collections business, addressing challenges such as email contact optimization, collection offer assignment, account charge-off risk prediction, and third-party debt settlement company enrollment prediction.• Software Engineering: Designed a data and model monitoring pipeline with a corresponding dashboard; implemented end-to-end Kubeflow training and re-fit pipelines; enhanced model algorithm runtime with batch processing and distributed computation.• Extracurricular Contribution: Organized a cross-team Data Science ML modeling lightning talk series.
EDUCATION
Columbia University
Master's degree, Management Science and Engineering
清华大学经济管理学院 Tsinghua University School of Economics and Management
Bachelor of Business Administration - BBA, Business Administration and Management, General
Columbia University
No Degree, Finance, General
Tsinghua University
Bachelor of Engineering - BE, Construction Management
ABOUT XIAOYI WANG
As a Data Scientist and Machine Learning Engineer, I specialize in developing innovative solutions that leverage advanced machine learning techniques, including reinforcement learning and large language models (LLMs). My experience spans the full ML lifecycle, from data ingestion and feature engineering to model development, deployment, and monitoring in production environments.
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.