Quantitative Analyst
Computer Code Programming
Computer Programming/CodingWorked on the end-to-end development and benchmarking of machine-learning credit risk models, involving evaluation of model outputs rather than manual annotation. Implemented and tested a refined decision tree and a challenger random forest regression model to compare consistent accuracy across SAS and Python. Contributed to feature engineering and exploratory analysis to support model segmentation for recoveries in the IRB Loss-Given-Default setting. • Benchmarked decision tree vs random forest regression accuracy using SAS and Python. • Conducted quantitative analysis to model Maximum Recovery Period and Independence Period from transactional data. • Performed EDA and feature engineering to identify influential variables for segmentation. • Supported cloud migration by testing the analytics environment on AWS.