Curriculum vitae
Babaniyi
Olaniyi
9M+insured people represented
100M+healthcare claims processed
€11M+annual pricing leakage protected
13%gross-profit improvement supported
Experience
01Data Scientist
Causal Foundry
- Built claims-driven healthcare-financing models used in policy decision-making, supporting modeled liquidity for health centres serving 9M insured people.
- Designed provider-performance, utilization, patient-outcome, and health-financing decision workflows across more than 100M claims.
- Reconciled 5,000+ drug SKUs across seven agencies using text similarity and embedding-based matching.
- Developed anomaly detection and automated indicators that improved oversight across 1,000+ providers.
Senior Data Scientist
ZF Group
- Designed pricing-arbitrage and anomaly-detection systems that identified inconsistencies and protected approximately €11M annually.
- Built elasticity and uplift models across a €500M+ portfolio, contributing to a 13% gross-profit improvement.
- Productionised PySpark and Databricks pipelines with engineering partners for commercial decision workflows.
Data Scientist
Benshi AI
- Forecast pharmacy demand using time-series and deep-learning methods, helping reduce stockouts by 18%.
- Improved mobile-health engagement with survival models, churn prediction, return models, and behavioural experiments.
- Designed adaptive interventions and recommendations using contextual bandits and reinforcement-learning concepts.
- Mentored data scientists and translated technical results for product, research, and policy stakeholders.
Data & Business Intelligence Analyst
The Alchemist Atelier
- Built segmentation and churn models to improve communication, retention, and customer satisfaction.
- Combined multiple data sources into revenue, budget, expense, and KPI dashboards that reduced manual reporting.
- Analysed acquisition, customer journeys, pricing scenarios, and commercial performance for management decisions.
Research Assistant
IAE-CSIC
- Built newspaper-text collection and predictive-modeling workflows to study political violence and conflict risk.
Research
02Data-Driven Approach to Capitation Reform in Rwanda
Claims-data-driven capitation design, calibration, monitoring, and quality insights.
Enhancing Product Recommendations with Multi-Modal LLMs
ISIR-eCom research on text and image signals for next-item prediction.
User Engagement in Mobile Health Applications
KDD work on probabilistic and survival-analysis approaches to engagement and churn.
Power Samade distribution: properties and application to real lifetime data
Nigerian Journal of Science and Environment.
Selected work