Jan 2019 — Present
Senior Data Scientist
ComScore
- Managed the end-to-end delivery of a core data product built on Python and Snowflake, implementing over
- 50 logic enhancements while utilizing CI/CD practices to ensure robustness and reliability across 10+
- internal teams.
- Developed a Difference-in-Differences analysis framework packaged into an automated production pipeline,
- enabling seamless integration with existing data workflows across 20 countries. This allowed for data-driven
- decisions that significantly enhanced the reliability of findings.
- Built an XGBoost churn model to identify at-risk panelists, operationalizing it within the production
- environment and reducing month-over-month variance by 5%, leading to 20% better resource allocation
- through proactive retention strategies.
- Collaborated with Product and Engineering teams to maintain an Isolation Forest anomaly detection model
- using Python, streamlining deployment and version control through Git, which reduced fraud incidents by
- 5% and yielded operational cost savings.