Sr. Data Scientist

About the role

The Sr. Data Scientist works remotely to design, build and operationalize advanced machine‑learning and GLM pricing models for property‑and‑casualty insurance products. Partnering with actuarial, underwriting and product teams, the role translates business problems into analytical solutions, validates model performance, and ensures governance for regulatory review.

About the company

Promantis Inc, founded in 2007, provides consulting, technology and outsourcing services across health, insurance, telecom, retail, banking and government sectors. It offers contract consulting, staff augmentation, enterprise and software application solutions, and systems/network engineering, leveraging a large expert staff to deliver customized, scalable solutions for global clients.

Requirements

• Based in the United States (Cincinnati, Ohio preferred)
• 7+ years experience in P&C insurance analytics, pricing or actuarial‑adjacent data science roles with advanced ML, NLP and deep‑learning techniques
• Hands‑on end‑to‑end ownership from data preparation to model deployment
• Proven ability to work with large, complex insurance datasets and explain results to non‑technical stakeholders
• Strong understanding of P&C insurance pricing concepts, customer life‑cycle, rating variables and risk segmentation
• Bachelor’s or Master’s degree in data science, economics, mathematics, computer science/engineering, operations research or related field
• Experience with machine‑learning algorithms for tabular data such as Gradient Boosting, Random Forests, XGBoost, LightGBM and NLP on unstructured data
• Expertise in GLM modeling (Poisson, Gamma, Tweedie, Logistic)
• Proficient in Python (pandas, numpy, scikit‑learn, statsmodels) for data analysis and modeling
• Skilled in Databricks/Spark (PySpark) for large‑scale data transformation and feature engineering
• Strong SQL skills for data extraction, transformation and analytical queries
• Knowledge of model explainability techniques (e.g., SHAP, partial dependence)
• Experience deploying models, building scoring pipelines and monitoring performance