The mission:
The Senior Credit Risk Data Scientist at Baubap is responsible for designing, implementing, and improving predictive models that directly impact our credit decisions and portfolio performance. This role will play a critical part in generating accurate forecasts, building data-driven methodologies, and continuously iterating based on real-world learning—always grounded in a deep understanding of the business context.
The expected outcome:
- Develop, deploy, and maintain machine learning models that improve the accuracy of forecasts across key business levers such as approval rate, disbursement rate, loss rate, and average loan amount.
- Deliver short-, mid-, and long-term forecasts for key portfolio and business indicators to guide strategic decision-making.
- Continuously improve models and data pipelines based on new insights, feedback loops, and shifts in portfolio dynamics.
- Build methodologies that reflect a clear understanding of the business and customer behavior—combining data science best practices with practical, real-world constraints.
The day to day tasks
- Model development & deployment: Design and implement predictive models (classification, regression, time series, etc.) to forecast credit risk metrics and optimize decision-making.
- Model iteration & lifecycle management: Regularly retrain and improve models based on recent performance, business evolution, and new data availability.
- Forecasting: Build robust models to predict portfolio KPIs over different time horizons (daily/weekly/monthly), including loss rate, disbursed amount, average ticket size, and approval rate.
- Experimentation: Collaborate with cross-functional teams to design and evaluate A/B tests or quasi-experiments that inform modeling improvements.
- Feature engineering: Create high-quality, interpretable features from raw transactional and behavioral data.
- Data exploration & root-cause analysis: Use statistical techniques to detect anomalies, understand shifts in model performance, and identify risks or opportunities.
- Business alignment: Partner closely with Risk, Product, Finance, and Data Engineering teams to ensure that models and methodologies are aligned with business goals and operational realities.
- Documentation & reproducibility: Maintain clear documentation of models, assumptions, and decisions to ensure transparency, auditability, and future scaling.
Why YOU should apply:
- 5+ years of experience developing, validating, and deploying predictive models in a production environment, preferably within financial services or credit risk.
- Strong proficiency in SQL for data extraction and transformation.
- Advanced skills in at least one programming language commonly used in data science, such as Python or R.
- Proven ability to build and tune machine learning models (e.g., classification, regression, time series forecasting), using libraries such as scikit-learn, XGBoost, LightGBM, etc.
- Experience maintaining and iterating on models based on real-world performance and shifting data patterns.
- Comfortable working with experimentation frameworks, A/B testing, and validation pipelines.
- Ability to translate complex technical insights into clear business recommendations.
- Strong understanding of statistical concepts and their application in risk modeling.
- Fluent in English (written and spoken); able to work and communicate effectively with an international and cross-functional team.
- Bonus: experience in financial risk areas
- Bonus: experience working with version control tools (e.g., Git), workflow managers (e.g., Airflow), and cloud-based data platforms (e.g., AWS, GCP, or similar).
What we can offer:
- Being part of a multicultural, highly driven team of professionals
- 20 vacation days / year + 75% holiday bonus (Prima Vacacional)
- 1 month (proportional) of Christmas bonus (Aguinaldo)
- Food vouchers
- Health & Life insurance
- Competitive salary
