Summary
Xiaodan Fang is a data scientist and engineering leader with nine years of experience building and validating ML models for high-stakes production systems, currently working on GTrade and Display Ads Auction at Google. She has a strong track record at Capital One leading model development and validation efforts, mentoring teams, and delivering measurable business impact—projects that include an XGBoost credit risk model with an estimated $50M annual NPV improvement and a credit line optimization that added ~$3M NPV. Skilled in scalable model validation, residual modeling, and automation, she created a Python package adopted into monitoring workflows and authored a comprehensive model validation guidebook for her team. Xiaodan combines deep academic training in operations research from Tsinghua and Cornell with hands-on experience in production ML, and has tackled large-data engineering challenges (e.g., speeding unsupervised pipelines 70x) that reveal a practical focus on robustness and operational efficiency.
9 years of coding experience
6 years of employment as a software developer
Management and Technology, Management and Technology at Technical University Munich
Cornell University
Bachelor's degree, Bachelor's degree at Tsinghua University