Data scientist & economist / Barcelona
Babs Olaniyi.
Models, experiments, and the decisions behind them.
I am a data scientist at Causal Foundry. My work brings together economics, statistics, and machine learning, from healthcare financing in Rwanda to pricing and recommendation systems.
I like working through the whole problem: what to measure, how to test it, and how to turn the result into something people can use.
Selected work
Full portfolioFood-crisis early warning from bilingual news
English and Arabic news turned into an auditable review queue for food-security analysts across five Mashriq countries.
Claims-driven health financing
Claims analytics, capitation reform, and provider monitoring using data representing more than 9 million insured people.
Finding and preventing pricing leakage
Anomaly detection, elasticity, and uplift models across a €500M+ industrial product portfolio.
Product recommendations with multimodal models
Combining reviews, images, and temporal signals for next-item prediction across approximately 3 million interactions.
Papers & publications
All publicationsData-Driven Approach to Capitation Reform in Rwanda
Health financing, claims data, and provider payment.
Enhancing Product Recommendations with Multi-Modal LLMs
ISIR-eCom / Text and image signals for recommendations.
User Engagement in Mobile Health Applications
KDD / Probabilistic and survival models of engagement.
Writing
All writingFavourite reads
A few bookmarksIdeas I return to when thinking about experiments, recommendations, and building software.
- Nobody Gets Promoted for SimplicityTerrible Software / Engineering
- Bandits for Recommender SystemsEugene Yan / Experimentation
- Understanding Multimodal LLMsSebastian Raschka / Machine learning
- Data Distribution Shifts and MonitoringChip Huyen / ML systems
- Research Software Engineering in PythonResearch practice
- Multi-Armed Bandits and the Stitch Fix Experimentation PlatformStitch Fix / Experimentation