Dawei Pang Email & Phone Number

San Francisco Bay Area

• Automated the ETL data pipeline in order to streamline 50GB+ of multi-modal traffic data with Python and Crontab
• Forecasted the congestion rate by using XGBoost, Decision Tree, random forest and Deep Learning models (CNN), and optimized the model, reduced the MSE for 5%
• Developed and maintained the real-time tableau dashboard to show the discrepancy between the forecasted rate and real-time congestion condition
• Identified the root cause of the discrepancy in 4+ different dimensions including seasonality, traffic zone and road condition etc., successfully improved the data reporting accuracy for 20%

• Cross-Sell Marketing Analytics
• Engineered 180K+ users’ data from database and extracted 15+ usable features including age, city, behavior etc.
• Performed K-means clustering algorithms in order to conduct customer segmentation and identified the target cohort
• Identified the key customer churn point following AARRR model for 5 cohorts and compared the difference
• Optimized the targeted coupon campaigns by feeding customized content to 5 customer clusters, achieved 70% boost for performancePlatform Layout Experimentation Analysis
• Analyzed A/B testing results in hypothesis testing and extracted causality between UI change and CTR

Chengdu, Sichuan, China

• Created the database for temperature sensors data and built the pipeline to process 300GB+ data in Python and MySQL
• Performed anomalous detection algorithms such as isolation forest and confidence interval for temperature data to identify the outlier
• Embedded the outlier threshold into the pipeline and set up the alert for data outlier issues, successfully improvednthe data accuracy for 30%
• Developed 4+ dashboards in Tableau served for 30+ stakeholders including trend analyses and anomaly detection, reduced the manual reporting time by 10 hours per week

Chengdu, Sichuan, China

• Defined 5+ core metrics in order to track the performance of the supply chain including inventory turnover, inventory accuracy, order fill rate etc.
• Forecasted the demands of the products by using linear regression and XGBoost model in Python, reached R2=0.89
• Optimized the supply chain based on the forecasted demands, reduced 12% reduction in raw material cost

Master'S Degree, Data Science

Bachelor'S Degree, Applied Mathematics