Ziyad Maknojia

At REGIONS BANK, I developed and deployed a real-time XGBoost model to detect high-risk check deposits, improving hold precision by 65% while minimizing customer friction.

I built production machine-learning workflows for credit-risk models, including a scikit-learn pipeline that takes models from raw data through feature calculation, encoding, and scoring. I also supported the transition of models into production using Snowflake, AWS Sagemaker, CI/CD pipelines, and DAG-based monitoring.

As an AVP – Risk Quantitative Analyst, I owned Allowance for Credit Losses governance and execution for the Consumer Credit Card portfolio. I implemented its CECL loss-forecasting model in SAS, automated forecasting workflows to cut execution time by 40%, and presented results and recommendations to senior leadership.

Earlier, as a Risk Credit Reporting Analyst, I automated PySpark and Python pipelines for regulatory and risk reporting and built Power BI and Tableau dashboards for CECL and CCAR metrics. I hold an M.S. in Business Analytics from George Washington University, where my graduate project used H2O AutoML to build and evaluate predictive models.