Data Science Intern

W. R. Berkley Corporation founded in 1967 is one of the nations premier commercial lines of property and casualty insurance providers. Each of the operating units in the Berkley group participates in a niche market requiring specialized knowledge about a territory or product. Our competitive advantage lies in our long-term strategy of decentralized operations allowing each of our units to identify and respond quickly and effectively.

Company URL: company is an equal opportunity employer.

We are seeking a highly motivated Data Science Intern who is interested in exploring data experimenting with modeling approaches and developing analytical solutions that influence business decisions. You will support the Advanced Analytics team by performing exploratory data analysis (EDA) data engineering feature engineering and statistical modeling to identify prioritize and explore predictive features for insurance loss modeling. The role will involve collaboration with stakeholders from the underwriting pricing and operations departments to ensure that data products align with business needs.

• Assess the business value of a third-party data source using an established scoring rubric
• Learn core insurance concepts and build foundational knowledge of Property & Casualty insurance products markets and underwriting processes
• Conduct exploratory data analysis (EDA) on insurance and oil & gas datasets to understand patterns data quality/coverage and modeling opportunities
• Explore data engineering approaches and test methods for entity matching including geospatial and temporal linkage techniques
• Engineer features by creating validating and maintaining derived variables that support analytical and predictive modeling
• Collaborate with business and technology partners to clarify requirements understand key variables and align on acceptance criteria
• Prototype predictive models and compare lightweight models against existing benchmarks
• Quantify value by estimating cost/benefit and return on investment for proposed features models and data products
• Develop analytics workflows and implement repeatable processes for EDA feature creation and model evaluation
• Plan model integration by working with engineering teams to outline how analytical models will be incorporated into production systems and reporting tools
• Create analytical datasets and dashboards to prepare datasets and build reports that help stakeholders make informed decisions
• Ensure data and model quality by conducting QA checks on datasets metrics and prototype models
• Maintain documentation for datasets features models and analytical processes
• Automate repetitive tasks using scripting and AI tools to streamline work and improve productivity

• Previous quantitative research experience (statistics applied mathematics computer science or related field) in a course personal project or previous role
• Experience using databases and data modeling concepts with proficiency in SQL (cloud data platforms a plus)
• Python proficiency strongly preferred (pandas NumPy scikit-learn; notebooks)
• Exploratory data analysis skills (profiling missingness outlier detection) and feature engineering experience
• Exposure to loss modeling pricing analytics or risk scoring is a plus
• Oil & Gas or Insurance domain knowledge is a plus (e.g. geospatial data underwriting context)
• Familiarity with Agile/SDLC practices is a plus

Education Requirement

• Current masters student or recent graduate in Data Science Analytics Statistics Engineering Computer Science or a related field is a plus