Data Scientist

Job description

**Role Overview:** We are seeking two skilled Data Scientists to join our engineering team to develop and implement a high-impact AI/ML predictive modeling solution. In this role, you will analyze historical claims data to generate VIN-level risk scoring and loss ratio forecasts for the Extended Service Business (ESB) Core Segment. Your work will directly influence product re-pricing and claim adjudication strategies, with a targeted objective of $1 million in annual claim cost reductions. **Key Responsibilities** 1. **Model Development** (Requires 35 years of experience): Design, build, and deploy end-to-end machine learning models on Google Cloud Platform (GCP) to predict loss ratios and identify high-utilization contracts at the VIN level. 2. **Data Engineering** & Feature Creation: Ingest and transform high-volume historical claims data from OWS and enterprise source systems; engineer robust features that capture granular risk factors. 3. **Advanced Analytics**: Apply statistical modeling and ML algorithms (regression, gradient boosting, etc.) to drive proactive risk assessment and dynamic product pricing. 4. **Stakeholder Collaboration:** Partner with business analysts and product teams to translate model outputs into actionable financial forecasting and strategic claim adjudication workflows. 5. **Operationalization**: Ensure model scalability and performance monitoring, contributing to a framework designed for future expansion across global markets. **Required Technical Skills** 1. Cloud Proficiency: Hands-on experience with **Google Cloud Platform (GCP)specifically BigQuery, Vertex AI, and Cloud Composer.** 2. Core Data Science: **Strong proficiency in Python/PySpark for data manipulation and model building (Scikit-Learn, XGBoost, LightGBM, or TensorFlow).** 3. Database Expertise: Advanced **SQL skills** for complex data retrieval, aggregation, and performance optimization within large datasets. 4. Model Lifecycle: Demonstrated experience in the **full ML lifecycle:** data cleaning, feature engineering, model training, validation, and deployment (MLOps).