Data Scientist III

Role overview

Data Scientist III at Hagerty, a hands-on “build-and-ship” role on the Data Science team. You’ll build the customer identity and personalization layer that helps Hagerty understand and engage members across its subscription and property & casualty (P&C) insurance products.

Responsibilities

Customer Identity & Data Foundations

• Build identity resolution across first-party and third-party data sources (member, household, vehicle, and behavioral signals from insurance and subscription touchpoints).
• Develop matching systems combining deterministic methods with probabilistic matching at scale, balancing precision/recall/cost.
• Partner with Data Engineering and the Customer Data Platform (CDP) to land resolved identities and audiences into production pipelines and activation systems.
• Evolve the identity layer toward graph-based representations of members, vehicles, and relationships.

Recommendation & Personalization

• Design, build, and evaluate recommendation/personalization models (content-based and hybrid) for next-best-product/content across insurance and subscription offerings.
• Build cold-start strategies for new and low-engagement members.
• Make trade-offs between real-time vs batch serving while accounting for latency and freshness constraints.

Predictive Modeling

• Develop well-calibrated predictive models for lifecycle behavior (e.g., churn/retention, propensity to buy, propensity to lapse/renew).
• Build next-best-action/journey-signal models to generate business-ready triggers for cross-sell and upsell.
• Own modeling workflows: exploratory analysis, feature engineering, model development, cross-validation, and performance monitoring.

Productionization & Collaboration

• Ship models as reliable production services with ML Ops (containerized deployments, automated testing, monitoring).
• Source/analyze features from Snowflake, SQL Server, and AWS RDS Postgres; collaborate with Data Engineering to scale features into pipelines.
• Contribute to modeling standards with maintainable, documented, testable code.
• Communicate methods, results, and trade-offs to technical and non-technical partners.

Requirements (from the posting)

• Hands-on experience designing, training, and deploying ML models in production.
• Proficiency in Python and modern ML frameworks: scikit-learn and XGBoost.

Nice-to-haves

• Not fully visible in the provided text (posting cuts off after “Strong …”).