Data Scientist Dallas - TX - Texas

Python with rating knowledge

7+

Data Manipulation: You must be highly proficient in Pandas and NumPy to clean, sort, and process large historical datasets.Machine Learning: Familiarity with Scikit-Learn is essential for building classification and regression algorithms that predict risk.Statistical Modeling: Use packages like Statsmodels to apply Generalized Linear Models (GLMs), which are the industry standard for insurance ratemaking.Version Control: Standard team workflows require the use of Git to manage code updates and track model versions securely.

remium Calculation: Using age, health status, and vehicle data to price policies and set premiums.Underwriting: Assessing the statistical risk of insuring an individual or business.

Key Roles & ResponsibilitiesRating Engine Development: Write and maintain backend code that ingests risk attributes and calculates accurate policy premiums, discounts, and surcharges.Actuarial Translation: Collaborate with actuaries to translate manual rate manuals (like those filed in SERFF) and statistical models into executable, production-grade code.Data Pipelines & ETL: Build and manage data pipelines using libraries like Pandas and NumPy to process historical claims and policy data.Predictive Pricing Modeling: Develop machine learning algorithms (e.g., using Scikit-Learn) to evaluate risk loss and optimize pricing models.Compliance & Auditing: Ensure rating logic complies with state insurance regulations by building logging and auditing mechanisms directly into the code.Core Technical SkillsProgramming Languages: Advanced proficiency in Python and SQL.Python Libraries: Pandas and NumPy for data manipulation Scikit-Learn for predictive modeling.Insurance Platforms: Familiarity with modern underwriting and actuarial platforms like Guidewire, hx Renew (hyperexponential), or Openkoda.Cloud & DevOps: AWS (Lambda, S3) or Azure services, Docker, and CICD tools.Version Control: Git GitHub for collaborative software development.Domain-Specific KnowledgeRatemaking Fundamentals: Understanding of loss cost modeling, frequency vs. severity distributions, and base rate calculations.Underwriting Rules: Knowledge of how Motor Vehicle Records (MVR), garaging territories, and vehicle safety features impact risk tiering.Telematics: Experience parsing and utilizing data from usage-based insurance (UBI) trackers to adjust rates based on driving behavior.

SQL (Structured Query Language): The coding language used to pull raw data from massive insurance databases.Predictive Modeling (Machine Learning): Using code to guess which drivers will cost the company the most money.Data Visualization: Using Python packages like Matplotlib or Seaborn to turn complex pricing data into easy-to-read charts for business leaders.Cloud Computing (AWSAzure): Running massive pricing models on remote computers so your laptop does not crash.Regulatory Compliance: Understanding state laws and rules to ensure your Python pricing models do not violate fair housing or discrimination rules.

Role Descriptions: Key Roles & ResponsibilitiesRating Engine Development: Write and maintain backend code that ingests risk attributes and calculates accurate policy premiums discounts and surcharges.Actuarial Translation: Collaborate with actuaries to translate manual rate manuals (like those filed in SERFF) and statistical models into executable production-grade code.Data Pipelines & ETL: Build and manage data pipelines using libraries like Pandas and NumPy to process historical claims and policy data.Predictive Pricing Modeling: Develop machine learning algorithms (e.g. using Scikit-Learn) to evaluate risk loss and optimize pricing models.Compliance & Auditing: Ensure rating logic complies with state insurance regulations by building logging and auditing mechanisms directly into the code.Core Technical SkillsProgramming Languages: Advanced proficiency in Python and SQL.Python Libraries: Pandas and NumPy for data manipulation Scikit-Learn for predictive modeling.Insurance Platforms: Familiarity with modern underwriting and actuarial platforms like Guidewire hx Renew (hyperexponential) or Openkoda.Cloud & DevOps: AWS (Lambda S3) or Azure services Docker and CICD tools.Version Control: Git GitHub for collaborative software development.Domain-Specific KnowledgeRatemaking Fundamentals: Understanding of loss cost modeling frequency vs. severity distributions and base rate calculations.Underwriting Rules: Knowledge of how Motor Vehicle Records (MVR) garaging territories and vehicle safety features impact risk tiering.Telematics: Experience parsing and utilizing data from usage-based insurance (UBI) trackers to adjust rates based on driving behavior.

Essential Skills: Data Manipulation: You must be highly proficient in Pandas and NumPy to clean sort and process large historical datasets.Machine Learning: Familiarity with Scikit-Learn is essential for building classification and regression algorithms that predict risk.Statistical Modeling: Use packages like Statsmodels to apply Generalized Linear Models (GLMs) which are the industry standard for insurance ratemaking.Version Control: Standard team workflows require the use of Git to manage code updates and track model versions securely.remium Calculation: Using age health status and vehicle data to price policies and set premiums.Underwriting: Assessing the statistical risk of insuring an individual or business.SQL (Structured Query Language): The coding language used to pull raw data from massive insurance databases.Predictive Modeling (Machine Learning): Using code to guess which drivers will cost the company the most money.Data Visualization: Using Python packages like Matplotlib or Seaborn to turn complex pricing data into easy-to-read charts for business leaders.Cloud Computing (AWSAzure): Running massive pricing models on remote computers so your laptop does not crash.Regulatory Compliance: Understanding state laws and rules to ensure your Python pricing models do not violate fair housing or discrimination rules.

Desirable Skills:

Keyword: