Carolina Contreras Portela

• Developed and structured databases and reporting frameworks to enhance automation and to support business objectives.
• Leverage analytics and visualization tools to design and present information to drive fact-based decision making.
• Supports cross-functional collaboration to drive effective data solution designs that comply with current and newly defined processes.
• Perform different statistical analyses, sampling and tests; identify outliers in data sets and conducting different segmentation studies.
• Provides documentation of coding and content to satisfy business and system requirements and reviews to meet compliance requirements
• Mentored junior team members, fostering professional growth and improved performance within the department.
• Managed multiple concurrent projects while maintaining attention to detail and prioritizing tasks effectively under tight deadlines.
• Improved reporting accuracy by streamlining data collection and analysis processes.
• Organized and updated databases, records and other information resources.

• Develop statistical and machine learning models to assess and predict risks, such as claims frequency, credit scoring, and loss estimation. Apply techniques like Bayesian inference, stochastic modeling, and Monte Carlo simulations.
• Extract, clean, and transform large datasets from various sources (SQL, APIs, CSV). Perform exploratory data analysis (EDA) and create meaningful features to enhance model performance.
• Build automated data pipelines in Python for actuarial calculations. Develop interactive reports and visualizations using Matplotlib, Seaborn, and Plotly, or dashboards with Power BI.
• Conduct backtesting models over technical provisions, reserves (e.g., IBNR), and solvency metrics. Simulate risk scenarios to estimate Value at Risk (VaR) and other financial indicators for capital management.
• Evaluate and validate predictive models using performance metrics (RMSE, AUC-ROC). Ensure compliance with regulatory frameworks and implement explainability techniques like SHAP and LIME.
• Support risk management and decision-making by translating complex actuarial models into actionable insights for stakeholders, ensuring alignment with industry regulations and business objectives.

Project: LAKLEMS & National Accounting Harmonization

• Integrate, clean, and analyze data from national surveys. Apply imputation techniques for missing data and identify trends in economic activity.
• Develop econometric models to estimate unobservable variables. Use time series methods (ARIMA, Kalman Filter) and machine learning techniques to improve accuracy.
• Implement growth decomposition models to measure the contribution of capital, labor, and technology to productivity. Under the LAKLEMS framework.
• Perform seasonal and calendar adjustments to ensure data consistency over time. Use techniques like X-13ARIMA-SEATS to standardize economic indicators.
• Build Python pipelines to automate TFP calculations and optimize workflows for large-scale economic data processing.
• Develop interactive dashboards and generate visual reports using Matplotlib, Seaborn, and Plotly to communicate key economic insights.
• ssess model accuracy with metrics like MAPE, RMSE, and R². Ensure transparency, reproducibility, and compliance with national accounting standards.

• Jun 2018 - Jul 2018: Supported the development of materials for the Summer School in Actuarial and Financial Sciences
• Feb 2018 - Jun 2018 | Mar 2017 - Nov 2017: Tutored introductory mathematics courses across multiple faculties