●​ Developed and deployed ML models for risk prediction and credit analysis, improving loan approval accuracy by 12%.

●​ Analyzed large financial datasets using Python and SQL, reducing data processing time by 30%.

●​ Applied linear and logistic regression models to analyze customer transaction behavior and predict credit risk scores, enhancing

fraud detection capabilities and reducing false positives by 18%.

●​ Built dashboard reports in Power BI to monitor portfolio performance and flag high-risk accounts, accelerating decision-making for

senior analysts.

●​ Collaborated with cross-functional teams (business analysts, engineers) to translate business requirements into data solutions.

●​ Deployed predictive models to production environments using Flask and Docker, ensuring seamless integration with Wells Fargo's

internal loan processing systems — reducing manual intervention by 20%.

●​ Built and deployed ML models for credit risk prediction using Flask ,and Docker within a CI/CD pipeline.

●​ Conducted A/B testing and model performance evaluations (using ROC-AUC, Precision, Recall) to validate and retrain models,

maintaining high model precision in fluctuating economic conditions.

●​ Delivered data-driven insights and presentations to senior leadership on portfolio performance trends, emerging fraud patterns,

and model explainability — empowering stakeholders to make faster, data-backed decisions.