Applied Economics researcher passionate about alleviating poverty through evidence-based development policy, with unique hybrid academic and professional expertise in computer science, data science and quantitative economics.
Led Experimentation/Causal Inference Team with the vision to enable fast and intelligent decision making by ramping features to targeted segments, assessing risk and measuring causal impact.
Grew team 4x and expanded platform from A/B testing to comprehensive causal inference platform incorporating quasi-experiments, AI models, targeting, segmentation/HTE, meta analysis, anomaly detection, notification, data visualization, etc.
Created industry-leading innovative features including automatic feature ramping, AI experiments, dynamic configuration, real-time targeting and monitoring.
Led performance and scalability initiatives for both online and offline infrastructure to maintain ultra low latency while scaling up traffic 10x. The platform became one of the largest online and offline platforms in the industry.
Data Analytics Engineer
Oracle
11.2010 - 02.2015
Designed and built cross channel (mobile, social, email) marketing platform with experimentation, personalization, targeting, scheduling, tracking and performance monitoring.
Built a rapid supply chain planning platform to detect anomalies, and compute metrics (supply, demand, inventory, sales, forecasting) with visualization for deep dive analysis (aggregation, drill downs, cross-sectional analysis).
Led company-wide platform scalability initiatives, reducing 80% of CPU time and 60% of memory footprint across marketing campaign backend services.
Education
Candidate For Master's of Economics - Data, Economics And Development Policy
Massachusetts Institute of Technology (MIT)
Cambridge, MA
Master, Heinz College of Information Systems And Public Policy - Computer Science and Data Science
Carnegie Mellon University
Pittsburgh, PA
Bachelor, College of Economics And Management - Economics, Business And Computer Science
Tongji University
Shanghai
Skills
Programming language: R, Python, Stata, Java, PL/SQL, Pig, Spark, C
Econometric analysis
Economic theories and development policy
Causal inference (RCT and quasi-Experiment)
Statistical modeling (network, time Series, high dimensional data)
Machine learning (clustering and classification, reinforcement learning, neural network)
Languages
English
Native or Bilingual
Chinese (Mandarin)
Native or Bilingual
Publications
KDD (Knowledge Discovery and Data) 2018 - SQR: Balancing Speed, Quality and Risk in Online Experiments
Unbiased Estimate of Causal Effects in Online Experiments, patented: July 2019
Unified Management Of Targeting Attributes In A/B Tests, patented: Sep 2018
Automatic Ramp-up of Controlled Experiments, patented: Sep 2017
Projects
Design and evaluate (quasi) experiments for social scientists: RCT, RDD, IV, Regression, OVB.
Network Effects High dimensional Data visualization, linear and non-linear clustering and classification, Gaussian Mixture EM.
Neural network and deep learning: feedforward, recurrent, convolutional networks. Natural Language Processing with reinforcement learning.
Social network analysis with graph centrality, spectral clustering and evolution.
Climate data time series analysis, (trend, seasonality, stationary) and stats modeling with ARMA models, Gaussian process and spatial prediction.
Morgan Stanley equity trade algorithm competition for High Frequency Trading in Carnegie Mellon University.