Work Experience
Data Scientist
EY · Dublin
Data Scientist at EY working between Predictive Analytics and Data Engineering. I lead the development and design of survival analysis models with full responsibility for modelling approach, covariate specification, and development.
Designed and deployed survival analysis models on right-censored data to forecast annual spend under a fixed, business-critical deadline. Validated using rolling-origin backtesting over multiple historical forecasts using only data available at each forecast date.
Built in-house custom Python package combining business rules, summary statistics and scenario simulation in one easy-to-use API. Used by junior analysts to speed up previously labour-intensive analysis.
Performed end-to-end development of a new dimension for a large data warehouse including EDA scoping, client requirements gathering, ETL stored procedure scripting, aggregation and transformation logic, and Airflow orchestration
Wrote a custom, production-ready data-lineage extraction tool in Python to map lineage across all pipelines feeding the data warehouse as the technology stack did not have this functionality natively. Pytested (70% coverage) including monkeypatch to spoof database calls. Gave new developers an intuitive map of a warehouse that previously took weeks of domain exposure to learn, and made pipeline dependency failures far simpler to trace