Leading analytical and modeling efforts to drive the continuous improvement of the AWS support engineering workforce management
Modeling contact forecasting, discovering insights and identifying opportunities through the use of statistics, machine learning, and deep learning to drive business and operational improvements
Building solutions that will help drive a more efficient operations network
Partnering with data engineering, tooling team, operations, training, capacity planning and finance teams, driving optimization and prediction solutions across the network
Data Science Manager
Capital One
01.2020 - 09.2021
Developed an AI-based vehicle auction sale price optimization system to maximize auto loan charge-off recoveries from auction sale of repossessed vehicles which increased the company's annual loss recovery by ~40 MM.
Improved the Account Level Valuation (ALV) system which predicts the expected present value of the future cash flows attributable to an individual auto loan.
Applied theoretical expertise and innovation to create or adopt new machine learning techniques to solve business problems.
Developed project roadmaps based on impact and effort, working with stakeholders to achieve short-term and long-term goals.
Provided coaching and mentoring to junior data scientists on statistical and machine learning techniques.
Sr. Operations Research Consultant
American Airlines
09.2016 - 01.2020
Developed a deep learning-based demand forecast system for the entire AA network.
Built an airfare purchase fraud detection model improving true positive rate from 40% to 85% and reducing annual revenue loss by 3MM.
Drove the development of AAdvantage loyalty program through consumer behavioral pattern analysis, impacting 56 MM customers.
Collaborated with the Citibank on building marketing campaign model for Citi/AAdvantage co-brand credit card acquisition.
Developed a travel recommendation system to provide personalized leisure trip offers to customers based their travel/online search history.
Sr. Financial Risk Modeler
Think Finance
11.2015 - 09.2016
Designed and developed statistical methodology for usage in: underwriting, existing customer management, marketing campaigns and collections.
Developed risk models using machine learning algorithms to minimize credit/fraud losses,maximize response and approval rates, and profitability of products.
Created statistical software packages (R)/macros (SAS) for financial risk modeling.
Created weekly and monthly reports to monitor model performance.
Designed and developed business logic, pricing strategies, business forecasts, while optimizing profitability.
Presented findings and made recommendations to Risk Management team and business leaders.
Built churn model for revenue management.
Sr. R&D Statistician
PepsiCo Inc
07.2014 - 11.2015
Delivered statistical experimental designs in support of Frito Lay North American R&D $472 MM annual new product growth plan.
Developed statistical methodologies for manufacturing process development and analytical chemistry.
Created global protocols for statistical practice across all PepsiCo R&D functions.
Provided consultation/training services to entire campus consisting of over 300 +associates.
Built statistical capabilities for sensory team in support of understanding consumer insights on PepsiCo's snack products.
Education
Master of Science - Computer Science
The University of Texas At Austin
05.2024
Ph.D. - Statistical Science
Southern Methodist University
Dallas, TX
2014
Bachelor of Science - Statistics
University of Science and Technology of China
Hefei, Anhui Province, China
2010
COMPETENCIES
Analytics: Association Rules, Bayesian Statistics, Bootstrap Aggregation (Bagging), Boosting, Deep Learning (Feedforward Neural Networks, Recurrent Neural Networks, Convolutional Neural Networks ), Design of Experiments, Elastic Net (LASSO, Ridge Regression), Ensemble Modeling, Factor Analysis, Generalized Additive Models, Generalized Linear Models, Generalized Procrustes Analysis, K-Nearest-Neighborhoods, Naïve Bayes, Natural Language Processing (NLP), Principal Component Analysis, Probabilistic Graphical Models, Partial Least Squares Regression, Regression/Classification Trees, Random Forest, Reinforcement Learning, Sampling Methods, Self-organizing Map, Statistical Process Control, Support Vector Machines, Survival Analysis,Time Series Analysis.
Languages/Tools: Python, R, SQL, Spark, Keras, TensorFlow, Tableau, SAS
PUBLICATIONS AND PRESENTATIONS
X. Wang, M. Chen, and O. Bai. “ABayesian Hierarchical Model for Meta-Analysis of Rare Binary Adverse Event Data.” Presented atJoint Statistical Meetings in Montreal, Canada Aug. 2013.
X. Wang, and O. Bai. “A Nonparametric Approach for Assessing TreatmentEffects Using Ranked Set Sampling with Randomized Block Designs.” Presented at Joint Statistical Meetings inBoston, MA Aug. 2014.
Bai, Ou, Min Chen, and Xinlei Wang. "Bayesian Estimation and Testing in Random Effects Meta-Analysis of Rare Binary Adverse Events."Statistics in biopharmaceutical research, no. 1 (2016): 49-59.
Li, Lie, Ou Bai, and Xinlei Wang. "An integrative shrinkage estimator for random-effects meta-analysis of rare binary events."Contemporary Clinical Trials Communications(2018).
PATENTS
Na Deng, Benjamin Segal, Ou Bai, Adam Thayer, Venkata Pilla. "Deep Learning-Based Demand Forecast System". No. 63/107,143, filed October 2020.
Ning Xu, Jose Antonio Ramirez-Hernandez, Steven James Oakley, Mei Zhang, Ou Bai, Supreet Reddy Mandala. "Predictive Sensor System For Aircraft Engines With Graphical User Interface." No. 62/770,035, filed November 2018.
Ou Bai, Wilfred Marcellien Bourg, Jr., Scott Fagan, Enrique Michel-Sanchez, Shahmeer Ali Mirza, Scott G. Richardson, Chen C. Shao. "Quantitative texture measurement apparatus and method." US9541537B1, granted January 10 2017.
Ou Bai, Wilfred Marcellien Bourg, Jr., Scott Fagan, Enrique Michel-Sanchez, Shahmeer Ali Mirza. "Feedback control of food texture system and method." US10070661, granted September 11 2018.
Ou Bai, Wilfred Marcellien Bourg, Jr., Scott Fagan, Enrique Michel-Sanchez, Shahmeer Ali Mirza, Scott G. Richardson, Chen C. Shao."Quantitative Liquid Texture Measurement Apparatus and Method." US10107785B2, granted October 23 2018.
Ou Bai, Wilfred Marcellien Bourg, Jr., Scott Fagan, Enrique Michel-Sanchez, Shahmeer Ali Mirza. "Quantitative In-Situ Texture Measurement Apparatus and Method." US20170176309A1, filed March 03 2017 and published June 22 2017.
Awards
Southern Methodist University
Research fellowship
Scheuren Award
John. E Walsh Award
Capital One Auto Finance
Diamond Award
Certifications
Practical Reinforcement Learning, December 2018, COURSERA
Introduction to Big Data, November 2015, COURSERA.
Introduction to Big Data Analytics, December 2015, COUSERA.
Hadoop Platform and Application Framework, December 2015, COUSERA.
Modern Design of Factorial Experiments, August 2014, American Statistical Association.
The Design and Analysis of Experiments that Use Computer Simulators, August 2014, American Statistical Association.
Classification and Regression Trees and Forests, August 2015, American Statistical Association.
Managing Statistical Consulting Projects, August 2015, American Statistical Association.
Statistical Issues in Online Experimentation, August 2015, American Statistical Association.
SAS Certified Base Programmer for SAS 9, May 2013, SAS, Inc.
SAS Certified Advanced Programmer for SAS 9, August 2013, SAS, Inc.
Environmental Program Manager (EPM) at Amazon Web Services (AWS), Amazon Data Services, Inc.Environmental Program Manager (EPM) at Amazon Web Services (AWS), Amazon Data Services, Inc.
Head - Enterprise Support (Mid and Large Enterprise Business) at Amazon Web Services (Amazon Internet Services Pvt. Ltd.)Head - Enterprise Support (Mid and Large Enterprise Business) at Amazon Web Services (Amazon Internet Services Pvt. Ltd.)
Customer Solutions Manager (Customer facing Program Manager) at Amazon Inc. (Strategic Accounts, Amazon Web Services)Customer Solutions Manager (Customer facing Program Manager) at Amazon Inc. (Strategic Accounts, Amazon Web Services)