1. Predicted FBA sellers’ lead time and restock frequency using Decision Trees algorithm, and improved the model accuracy2. Built data pipeline to automate the whole process of data prep, transfer, model training, testing, and predicting3. Provided business insights for supply chain decision making via big data analytics4. Established seller trust index by utilizing Random Forest and XGBoost algorithms to predict the acceptance rate of restock recommendations

1. Conducted advanced analytics for data-driven decision making to help clients optimize operation process and improve efficiency2. Developed predictive models and automate the personalized recommendation product, leveraging deep learning (RNN using Keras) to provide insights for customer behavior, marketing/sales strategy, and resources allocation3. Managed projects and collaborate with the cross-functional team to define problems, recognize technical challenges, and develop reliable solutions4. Experienced trainer for R Programming

1. Collaborated with environmental scientists to improve algorithms that identify statistical outliers in flood records2. Built the model to simulate flooding events caused by disparate physical processes, and offered improved risk analysis methods to hydrologists and industry engineers

1. Instructed students spending approx. 20 hours weekly teaching courses in Probability and Statistics, Risk Analysis and Management, and Explorations in Engineering2. Supervised 5 teaching assistants, overlooking grading standards, exam questions, test prep, and discussion session materials3. Developed lecture tutorial for R programming as a tool to help students better understand mathematical theories

1. Analyzed IT outsourcing demands in samll/medium business, offering insights and proposals to create a flexible market strategy for ITaaS (IT as a Service)2. Targeted 200 small business C-Suite executives with a comprehensive test survey focusing in 2 specific IT research categories: perspective and performance3. Developed and managed team blueprint to ensure that deadlines, logistics, and research meet client’s expectation

1. Evaluated adaptive methods for flood risk pertaining to climate change; presented the proceeding paper at the World Environmental and Water Resource Congress2. Collaborated with hydrologist and leading scholars to update the National Guidelines for Flood Frequency Analysis

1. Analyzed priority pollutants following EPA standard test methods to ensure water quality2. Worked closely with environmental scientists to exam water/soil samples from oil-spill contaminated sites; interpreted test results and indices to private clients and public agencies

1. Evaluated a 25-year plan of the Detroit combined sewer overflow (CSO) control program with budgets exceeding $800 million2. Recommended stormwater Best Management Practices (BMPs) based on cost-benefit analysis