Atlanta, Georgia5 days ago
li>A strong hands-on modeling background across both traditional techniques (logistic regression, scorecards) and modern ML (gradient-boosted trees such as XGBoost/LightGBM/GNN), with the judgment to know when each is appropriate and to hold a team to a high bar.
End-to-end modeling across the credit and fraud lifecycle — application/underwriting scoring, behavioral and account-management models, line and limit assignment, risk-based pricing, loss forecasting, and collections/recovery optimization.