Finance & Data Professional with 6+ years of experience. I combine FP&A expertise with SQL & Python to build automated forecasting models, reduce variance, and drive executive decision-making.
Forecasting across 16 business units was manual and prone to error, leading to inefficient capital allocation.
Consolidated $500M+ in annual Opex spend data. Designed SQL pipelines to process 500k+ annual transactions from legacy banking systems.
Replaced manual spreadsheets with automated driver-based models. Built Power BI dashboards to track variance in real-time.
Result: Improved forecast accuracy by 67% and shortened the planning cycle by 30%. Identified $5M+ in efficiency opportunities.
How can we accurately predict borrower default risk to minimize bad debt exposure?
Utilized historical borrower data, cleaning features related to income, debt-to-income ratio, and credit history.
Developed a classification model using Random Forest. Implemented feature importance scoring to understand key drivers of default.
Result: Improved default prediction accuracy by 18% over baseline models, enabling data-driven risk-based decisioning.
How can retailers optimize inventory levels during high-volatility holiday seasons?
Analyzed time-series sales data. Handled seasonality and holiday spikes using Prophet and XGBoost libraries.
Compared traditional time-series models against ML approaches to find the lowest error rate for peak periods.
Result: Reduced forecast error by 22%, supporting better inventory and staffing optimization.