I transform raw data into high-precision predictive systems. From deploying time-series models on the cloud to architecting business-critical BI dashboards, I build data solutions that protect ARR and optimize operational ROI.

I don't just build models. I build business answers.
Faith, family, and the outdoors keep me grounded. As a Christian, I lead with integrity. As a Barcelona fan, I appreciate teamwork and strategy. And whether it's hiking or spending time with loved ones, I recharge by staying active and connected.
Professionally, I'm a Data Scientist & Analytical Engineer focused on the full data lifecycle. I believe a model is only as good as the business problem it solves. My work bridges the gap between complex algorithms and executive-level decision making.
My core stack centers on Python (CatBoost, Scikit-Learn), SQL, and Cloud Infrastructure. I don't just deliver a notebook, I deliver a system whether that's a cloud-deployed forecasting API or a high-precision churn detection engine.
My work focuses on the intersection of Predictive Systems, Revenue Optimization, and Business Intelligence.
Bike-sharing platforms struggle to balance supply and demand, leading to missed revenue during surges and under-utilized assets during off-peak periods
Architected a production-grade MLOps pipeline to forecast city-wide demand with a 12-hours lead time, enabling data-driven dynamic pricing and optimized fleet distribution
Reduced forecasting error (MAE) by 51% over baseline. Developed the infrastructure to maximize revenue during peaks and stimulate demand during periods of low utilization
Telecom providers rely on Reactive Retention only identifying at-risk users after a cancellation request, leading to significant Annual Recurring Revenue (ARR) leakage.
Developed an end-to-end analytical system using CatBoost and Optuna to transition to a proactive model. Addressed a 73/27 class imbalance via minority upsampling.
Achieved 0.64 precision on churn risk, identifying the Tenure effect (first 12 months) as the primary window for high-ROI retention intervention.

Traditional credit models fail regulatory scrutiny due to hidden algorithmic bias.
Engineered a fairness-aware decision engine with Disparate Impact Audit pipeline to detect and mitigate bias.
90.74 RMSE accuracy, 100% Disparate Impact Audit pass
Python, SQL, AWS & Machine Learning
Professional Impact & Results
Data Scientist & Research Analyst at The Catholic University of America. Developed 'Churn Risk' models using Python to analyze retention. Engineered automated SQL cleaning workflows that reduced reporting errors by 95% and improved turnaround time by 20%.
Data Infrastructure Engineer at Women of Faith. Architected a centralized data warehouse from scratch using SQL Server. Replaced manual processes with scalable workflows, boosting operational efficiency by 25% and leading technical training for 20+ staff.
Education.subtitle
Education.educationDesc1
Education.educationDesc2
Education.educationDesc3
Let's turn your Data into Business Decisions