At CHRISTUS Colombia, I analyzed historical data, technical notes, and financial projections to identify contract over-execution and support ongoing negotiations. I also improved the traceability of 2,534 surgical package records using a data model in KNIME and Python.
For Compensar EPS through Grupo Proyéctame, I segmented insured members into 22 clinical and financial risk groups using classification and regression models. I designed a methodology for allocating the 2026 UPC budget by cohort and risk profile, projecting a 10% reduction in claims.
At Servicio Occidental de Salud (SOS), I identified a 10% recovery of financial resources across 72 health-model contracts. I also reduced medical materials costs by 15% and improved the analytics team’s efficiency by 30% through an item-homologation pipeline and a Power BI dashboard.
My work spans actuarial analytics, expected credit loss under IFRS 9, and machine learning across health and financial services. I use Python, SQL, R, KNIME, Power BI, Databricks, and MLflow to develop analyses and models that support financial and executive decisions.

