Staff Data Scientist / ML Engineer, LTV

Root is seeking a Staff Data Scientist to lead the development and technical direction of its customer lifetime value ecosystem. The role combines hands-on data science and production machine learning with model deployment, system health, business-impact measurement, and technical mentorship.

Responsibilities:

• Own the technical direction of LTV modeling. Frame ambiguous problems, weigh analytical options, make principled tradeoffs, and change course as new evidence emerges
• Improve the models that drive major decisions. Lead complex initiatives across interconnected conversion, retention, future-premium, and claim-loss models; understand how component models interact, diagnose underperformance, and prioritize improvements by business value
• Keep a complex codebase and production system healthy. Guide the design and evolution of the LTV codebase and its training, inference, and deployment workflows so they remain modular, tested, reliable, efficient, extensible, and readable as models and dependencies grow
• Stay hands-on across the lifecycle. Contribute directly to exploratory analysis, feature development, modeling, production Python, deployment, monitoring, and production support
• Measure what matters. Design experiments and validation frameworks with clear success criteria, then evaluate model performance and business impact after launch
• Help plan and deliver quarterly and long-term priorities. Partner with the team manager to sequence work around capacity, milestones, and dependencies; communicate risks early and help remove blockers
• Bridge modeling and engineering. Work across data science and machine learning engineering boundaries to ship reliable models, simulations, and forecasting workflows, and explain recommendations, risks, and tradeoffs to leaders across Marketing, Finance, Product, and Customer Experience
• Raise the bar for the team. Coach data scientists and machine learning engineers through code review, design feedback, and mentorship; build reusable tools and standards that improve work across Quantitative Science and strengthen an inclusive, collaborative team culture

Requirements:

• BS, MS, or PhD in Statistics, Computer Science, Economics, or a related quantitative field
• 8+ years of experience delivering complex, high-impact data science and production ML work
• Your background may be anchored in data science or machine learning engineering, but you bring considerable hands-on experience in both and deep expertise in developing models that drive business decisions
• Strong command of foundational data science principles, including statistical methods, predictive modeling, time-series forecasting, experimental design, measurement, and validation, with deep expertise in survival analysis, including time-to-event modeling, cure modeling, recurrent events, and censoring
• Deep expertise in Python and SQL, with strong software engineering discipline in Python, including writing modular, well-tested, maintainable Python with clear interfaces and type annotations, and experience maintaining and refactoring a large shared codebase over time
• Experience designing, deploying, and operating systems of interacting production models, such as ensembles or chained predictions, including shared training and inference pipelines, workflow orchestration, version control, CI/CD, monitoring, and production support
• Demonstrated ability to frame ambiguous modeling and systems problems, evaluate technical tradeoffs, estimate the value of the work, and prioritize high-leverage opportunities
• Strong communication and relationship-building skills, with the ability to connect technical work to business goals and explain decisions, risks, and tradeoffs to varied audiences
• A track record of shaping technical direction and raising the quality of others' work through coaching, review, and shared standards while remaining accountable for hands-on delivery
• Experience with customer lifetime value forecasting, simulation workflows, forecast-versus-actual analysis, or causal inference
• Experience with insurance or regulated financial products
• Experience with tools in or adjacent to Root's ML stack, such as AWS, Docker, dbt, Airflow, Metaflow, or MLflow
• Experience turning prototypes of new methods or data science tools into durable improvements in how a team works