- Developed an end-to-end systematic trading system for US equities using fundamental/macro/price data extracted from Compustat point-in-time snapshot dataset/CRSP/Fred API, created a library of +250 factors based on this dataset- Generated trading signals via machine learning, constructed and optimized a portfolio which rebalances quarterly- Created a backtesting engine looks at total return, drawdown, Sharpe ratio, concentration and tested robustness of strategy under different interest rates regimes

- Fintech startup affliated with Columbia University and City University of Hong Kong- Led feature engineering for a random forest model predicting Chinese bond issuers' default probability, model achieved ~70% accuracy in predicting defaulted Chinese bond issuers in 2021- Spearheaded data pipeline that extracted and cleaned financial data of 4,000+ Chinese bond issuers and macroeconomic data from Wind, Bloomberg and Refinitiv using Python and SQL- Built an interactive dashboard using PowerBI for default prediction model and presented to pitched to potential clients - Converted 3 clients, including HSBC and Refinitiv, generating the company's first revenue stream- Managed a team of 6 interns, overseeing their day-to-day tasks, providing guidance and ensure timely completion of projects

- Automated daily trade account reconciliation process with multiple brokers using python, javascript and SQL- Researched and presented algorithmic execution strategies (vwap, twap, pov, snipers etc.) to entire operations department

Conducted in-depth literature review on the use of machine learning in algorithmic trading for the Electronic Execution department, providing critical insights and knowledge to implement machine learning to existing execution strategies. Reported to the head of electronics execution department