Lead Data Scientist, Telematics

Root Insurance

17d•$142,800 - $178,500•Remote

About The Position

Root is seeking a Lead Data Scientist I to support the Telematics Data Science team. This is a hands-on role focused primarily on developing and maintaining telematics pricing and underwriting models, with a significant secondary responsibility for the data engineering and production systems that support those models. The role is expected to spend approximately 65% of its time on Data Science and 35% on Data Engineering. Data Science responsibilities include problem definition, dataset development, feature engineering, statistical modeling, model evaluation, stakeholder validation, regulatory support, and post-deployment monitoring. Data Engineering responsibilities include building and maintaining reliable modeling datasets, improving data pipelines and transformations, validating data quality and lineage, productionizing models, troubleshooting pipeline and system issues, and improving the reliability and maintainability of the Telematics data ecosystem. This role requires someone who can work across the full lifecycle of a telematics model—from raw data ingestion through model development, production deployment, monitoring, and ongoing operational support. The successful candidate will partner closely with Data Science, Data Engineering, Software Engineering, Pricing, Actuarial, Product, and Compliance teams to deliver reliable and measurable business outcomes.

Requirements

Strong knowledge of statistical modeling and machine learning methods, including GLMs, tree-based models, time-series analysis, feature engineering, model evaluation, resampling, and hyperparameter tuning.
Demonstrated experience building and maintaining production-quality data pipelines, modeling datasets, and data transformations.
Strong programming skills in Python and SQL, including the ability to write, test, review, and maintain production code.
Experience with version control, automated testing, code review, CI/CD, and cloud-based data or machine learning systems.
Experience validating data quality, lineage, completeness, consistency, and schema stability.
Ability to troubleshoot issues across data, application, pipeline, and infrastructure layers.
Experience deploying, monitoring, and maintaining production machine learning models.
Advanced degree in a quantitative discipline (PhD preferred) and/or 5+ years of applying advanced quantitative techniques to problems in industry.
Strong demonstrable knowledge of topics such as statistical modeling, machine learning, and numerical optimization.
Exceptional communicator and storyteller with strong data visualization skills.
Strong programming skills with experience in SQL and Python.
Demonstrated experience building, validating, and applying statistical machine learning methods to real world problems.
Demonstrates ownership mentality, taking initiative to find, prioritize, and be accountable for the highest impact work.
Ability to frame functional problem statements for the next 1-2 months, consistently making good decisions about the right path to follow in a well-defined problem space.
Experience using version control (e.g. Git) and cloud computing (AWS or analogous).

Nice To Haves

Familiarity with advanced methods such as neural networks, survival analysis, causal inference, or Bayesian modeling.
Experience in one or more insurance lines (P&C, Life, Health, Specialty).
Familiarity/understanding of insurance concepts such as Loss ratios, Loss Cost, Claims frequency etc.

Responsibilities

Develop and maintain telematics pricing and underwriting models.
Build and maintain reliable modeling datasets.
Improve data pipelines and transformations.
Validate data quality and lineage.
Productionize models.
Troubleshoot pipeline and system issues.
Improve the reliability and maintainability of the Telematics data ecosystem.
Work across the full lifecycle of a telematics model—from raw data ingestion through model development, production deployment, monitoring, and ongoing operational support.
Partner closely with Data Science, Data Engineering, Software Engineering, Pricing, Actuarial, Product, and Compliance teams.
Lead innovation by introducing new statistical algorithms and modeling techniques.
Prototype, test, retest and scale approaches.
Adopt and refine modeling best practices, documentation standards, and peer reviews.
Ensure statistical models comply with state-specific regulatory requirements and constraints.
Provide modeling support, documentation, and explainability to state regulators.