Senior Data Scientist – Risk Modeling (Senior Data Scientist – Modelado de Riesgos) - Hybrid

• Develop credit risk models: Design, build, validate, and maintain predictive models for credit origination, behavioral risk, portfolio management, and other risk use cases.
• Own the modeling lifecycle: Work across the full model lifecycle, including problem definition, population and target construction, feature engineering, model development, validation, backtesting, calibration, monitoring, and recalibration.
• Drive advanced risk analytics: Use SQL and Python to explore large datasets, identify portfolio trends, analyze delinquency and losses, and translate findings into actionable risk strategies.
• Strengthen credit decisioning: Support the development and optimization of underwriting strategies, score cutoffs, credit limits, segmentation, and portfolio management policies.
• Monitor model and portfolio performance: Build monitoring frameworks to track model discrimination, calibration, stability, data drift, portfolio trends, vintages, roll rates, delinquency, and other key risk indicators.
• Improve data and modeling quality: Validate data sources, implement data quality controls, assess feature stability, and identify potential issues such as leakage, selection bias, or population drift.
• Work with rejected and unobserved populations: Contribute to methodologies for addressing reject inference, selection bias, thin-file populations, and limited performance information where relevant.
• Develop in a modern ML environment: Use Databricks, MLflow, GitHub, Python, SQL, scikit-learn, and other appropriate modeling tools to build reproducible and well-documented analytical solutions.
• Support model implementation: Collaborate with Data and Engineering teams to ensure models developed by Risk Data Science can be reliably deployed and integrated into business decision flows.
• Translate analytics into business decisions: Communicate complex analytical findings clearly to Risk leadership and non-technical stakeholders and help turn model outputs into actionable business strategies.
• Contribute to Risk Analytics standards: Help build scalable methodologies for model development, validation, monitoring, documentation, and governance across Mexico, Brazil, and Colombia.

• 4–6+ years of experience in Data Science, Risk Analytics, Credit Risk, or related analytical roles.
• At least 2 years of hands-on experience developing or validating credit risk models or other predictive risk models.
• Strong proficiency in Python and SQL for data manipulation, statistical analysis, and model development.
• Experience working with Databricks or similar cloud-based analytics platforms.
• Experience developing predictive models using libraries such as scikit-learn, LightGBM/XGBoost, PyTorch, or equivalent tools.
• Understanding of the full model lifecycle, including development, validation, backtesting, monitoring, recalibration, and documentation.
• Strong understanding of credit risk analytics, including concepts such as:
  • delinquency and default;
  • vintage analysis;
  • roll rates;
  • bad rates;
  • portfolio performance;
  • score discrimination and calibration;
  • population and model stability.
• Experience working with large financial or transactional datasets and strong commitment to data quality and integrity.
• Ability to translate quantitative analysis into credit strategies and business recommendations.
• Working proficiency in English and Spanish.
• Academic background in Statistics, Mathematics, Economics, Engineering, Computer Science, Actuarial Science, Data Science, or a related quantitative field.
• Ability to work in a fast-moving environment and collaborate across Risk, Data, Engineering, and business teams.

• Experience in fintech, lending, credit cards, payments, or B2B financial products.
• Experience with Latin American credit markets, particularly Mexico, Brazil, or Colombia.
• Knowledge of credit bureau data and alternative data sources.
• Experience with PD modeling, expected loss, ECL, LGD, or EAD methodologies.
• Experience with reject inference or modeling under selection bias.
• Experience defining credit line strategies, cutoffs, risk segmentation, or underwriting policies.
• Experience with MLflow, model registries, version control, and reproducible ML workflows.
• Experience with Git and GitHub.
• Knowledge of data engineering concepts and ETL/data pipelines.
• Experience taking models from development through implementation in partnership with Engineering.
• Experience with visualization or BI tools such as Metabase.
• Master’s degree in Statistics, Data Science, Machine Learning, Economics, Finance, or a related quantitative field.

Where we work

This is a hybrid role: you'll split your time between the office, customers and home. We don't set a minimum number of office days for most roles — you'll spend time in the office naturally, and most days during your ramp-up or when your leader asks.