The position. 

We believe a large part of building an effective insurance company can be solved with a principled quantitative framework. We are committed to the rigorous development and effective deployment of modern statistical machine learning methods to problems in the insurance industry.

We are looking for Data Scientists to join our Lifetime Value (LTV) Analytics team. As part of this team, you will create models that predict conversion, retention, and/or loss cost. The team also builds and maintains tools that bring these models together to predict the lifetime value (and other related key metrics) of Root customers. You’ll work closely with partners across finance, marketing, pricing, product management, and other data science teams to support different business cases and optimize business outcomes.

We are hiring various levels of Data Scientists. Candidates with outstanding skills and experience will be considered for more senior-level roles.

What you’ll be doing.

• Applying principled methods to quantitative insurance challenges, estimating customer lifetime value
• Building data processing pipelines to quickly iterate on research ideas and put them into production
• Creating reports and dashboards displaying insights from models and monitoring model performance
• Effectively communicating insights from complex analyses
• Taking end-to-end ownership of problem domains and continuously improving upon quantitative solutions
• Leverage internal and external data sources to engineer novel features that enhance the accuracy of our models

What we’re looking for.

• Advanced degree in a quantitative discipline and/or 3+ years of applying advanced quantitative techniques to problems in industry
• Strong demonstrable knowledge of topics such as Bayesian statistics, machine learning, and numerical optimization
• Demonstrated experience applying statistical machine learning methods to enhance the decision making of an organization
• Exceptional communicator and storyteller
• Strong programming skills with experience using modern python or R packages. Experience with GBM libraries such as XGBoost or CatBoost is preferred
• Experience sourcing data for modeling from a star schema data warehousing architecture and feature engineering using SQL

Exceptional candidates may have experience in one or more of the following.

• Insurance rating plan development (personal lines auto insurance preferred)
• Insurance conversion and/or retention (churn) modeling experience
• Taking models to production and monitoring their performance
• Causal inference from observational data
• Generalized linear models, Tweedie distribution
• Using models to inform digital advertising or lead bidding decisions
• Experience using AWS cloud infrastructure (ec2, s3, Sagemaker)