<p>Architected enterprise-scale commercial insurance data ingestion pipelines using AWS and Snowflake, consolidating policyholder, quote, and claims data from 15+ source systems to enable unified underwriting analytics.</p><p> Designed and implemented Medallion Architecture (Bronze/Silver/Gold) in Snowflake and S3, organizing raw policy event data, curated risk features, and business-ready underwriting metrics, reducing time-to-insight for the actuarial and product teams by 60%.</p><p> Built event-driven streaming ingestion pipelines using Kafka and AWS Lambda to capture real-time policy bind, quote, and cancellation events from Foxden microservices, enabling live rate adjustment signals and reducing data latency from hours to under 3 minutes.</p><p> Developed 20+ analytics dashboards surfacing key insurtech KPIs, empowering underwriting, product, and executive teams to make data-driven pricing decisions.</p><p> Implemented CI/CD pipelines for data workflows using GitHub Actions and AWS CodePipeline, automating testing and deployment of Snowflake DBT models and Lambda-based ingestion jobs, reducing release cycle time by 65% and eliminating manual deployment errors.</p><p> Led data platform build-out to support Foxquilt’s US market expansion and Markel partnership, designing cross-border data schemas to unify Canadian and US policy data under a single Snowflake warehouse while maintaining regulatory separation between jurisdictions.</p><p> Developed DBT transformation models to standardize risk signal features used by Foxquilt’s ML underwriting algorithms, ensuring consistent feature definitions across the actuarial rating engine and analytics layer, and reducing data discrepancies in pricing decisions by 35%.</p><p> Designed and deployed Airflow DAGs on Astronomer to orchestrate nightly ingestion, DBT transformation, and data quality workflows across Foxden API sources, third-party partner feeds (Munich Re, Markel), and internal Snowflake layers.</p><p> Optimized AWS Lambda concurrency, S3 storage tiering, and Snowflake virtual warehouse auto-suspend settings, reducing cloud infrastructure spend by 30% while supporting a 3x increase in daily policy event volume driven by new partner onboarding.</p><p> Collaborated with Legal and Compliance teams to embed PIPEDA data privacy controls and FSRA regulatory requirements directly into pipeline design, ensuring policyholder PII was masked at ingestion, audit-logged at transformation, and retained per Ontario insurance data governance standards.</p>
