• Model Implementation: Utilized Learning to Rank model (LambdaMART) and Random Forest (RF for Rank) for cross-sectional analysis in evaluating Private Placement opportunities in the stock market, achieving above 0.7 Precision@K for K = 10.• Feature Selection: Integrated feature selection and categorization using methods such as RF for rank, VIF analysis and RFE (Recursive Feature Elimination), categorizing variables into core, control, and augmenting features.• Feature Development: Facilitated collaboration within the team to research, replicate, and construct new features, such as Sector Rotation and Restructured Fundamentals.

• Volatility Surface Modelling: Calibrated SABR model for option implied volatility in both Black and Bachelier spaces. Applied the Akima Interpolation Method between implied volatility points per tenor to construct the volatility surface.• Sensitivity Analysis: Conducted deterministic scenario analysis by adjusting parameters in the SABR model, providing preliminary analysis of risk through visualization and Greek letters.• Automated Data Pipeline: Developed and implemented an automation script using Python, SQL, and a third-party API to extract and cross-reference stock announcements in real time.

• Product Pricing: Utilized the DCF model to calculate pricing for cloud server and database products, considering cash flow potential under varying rebate policies, including basic, performance-based, and contra rebates.• Rebate Evaluation: Developed a dynamic forecasting model using simple logistic regression in Excel VBA, providing rolling updates on rebate fulfillment likelihood and estimating future liabilities, and offering reference to optimize rebate structures.