Data analytics professional and graduate researcher at Columbia University, specializing in translating complex datasets into strategic decisions through machine learning, business intelligence, and statistical modeling.
With experience at Capgemini and Shengtetai Food, I have built systems that directly impact operations and revenue — from automated supply chain pipelines to real-time dashboards tracking $1.5M in annual sales. My research on Bidirectional LSTM architectures for stock price forecasting was published at IEEE MLHMI 2024.
Skills
Education
Experience
- Analyzed Fortune 500 supply chain data from SAP using SQL and R, delivering strategic insights to C-suite leadership.
- Built Power BI dashboards for real-time visibility into supplier performance and procurement risk.
- Reduced manual data processing effort by 40% through automated workflows and UiPath–SAP integration.
- Engineered Python pipelines (Pandas, NumPy) processing 50,000+ rows of sales and inventory data.
- Developed Tableau dashboards providing leadership with real-time visibility into $1.5M in annual revenue.
- Identified high-value customer segments through SQL-based behavioral analysis, improving targeted marketing performance.
Research & Projects
Co-authored a peer-reviewed paper proposing a Bi-LSTM architecture that outperforms standard LSTM for time-series stock prediction across 11 stocks in 8 sectors. Developed an accompanying trading strategy with measurable alpha over a buy-and-hold baseline.
Engineered features from ad content, sentiment scores, and audience targeting metadata to train an XGBoost classifier. Reduced RMSE by 33% through hyperparameter tuning, regularization, and cross-validation.