Insurance Claim Data Scientist

Position Responsibilities

• Translate risk management business requirements into well-defined data science solutions, including incident prioritization and claim severity classification.
• Profile, clean, and prepare claims and incident data for analytics, modeling, and scoring.
• Develop feature engineering logic using structured and unstructured claims and incident data.
• Apply NLP and text-processing techniques to claim and incident narratives to extract useful risk signals.
• Develop record-linkage approaches to connect incidents and claims when a clean unique identifier is not available.
• Build and validate models that rank incidents by likelihood of becoming claims or requiring Risk Management intervention.
• Build and validate claim severity models that classify claims by likely financial impact and high-dollar claim risk.
• Generate explainability outputs, including key risk drivers and business-readable reasons for flagged incidents or claims.
• Collaborate with Risk Management, Legal, Data Engineering, BI, Data Governance, and MLOps partners to deliver usable business outputs.
• Monitor model performance, drift, scoring quality, and retraining needs.
• Document modeling assumptions, feature logic, validation results, limitations, and handoff requirements.
• Ensure data science work follows data governance expectations, including appropriate handling of PII and sensitive fields.
• Present findings, model results, and recommendations to business and technical stakeholders in a clear, actionable manner.

Experience and Qualifications

• Required Skills:
• Strong experience building supervised machine learning models, especially classification, ranking, and severity/risk scoring models.
• Strong experience with data profiling, data cleaning, feature engineering, model validation, and model evaluation.
• Experience working with messy, sparse, real-world enterprise datasets.
• Strong Python and SQL skills.
• Experience with NLP or text analytics, including narrative cleaning, text classification, embeddings, keyword extraction, or summarization.
• Experience with probabilistic record linkage, entity resolution, fuzzy matching, or deduplication.
• Experience explaining model outputs using feature importance, SHAP, reason codes, or other explainability methods.
• Experience working with Snowflake or similar enterprise data warehouse platforms.
• Experience supporting batch scoring, model monitoring, and production handoff.
• Strong understanding of data governance, sensitive data handling, and PII masking or exclusion.
• Excellent communication and teamwork skills.
• Preferred Skills:
• Experience with insurance claims, risk management analytics, litigation analytics, fraud detection, or safety analytics.
• Experience with claim severity modeling or high-dollar claim prediction.
• Experience with AWS, SageMaker, or similar cloud-based data science environments.
• Experience supporting BI outputs in Tableau, Power BI, Looker, or similar tools.
• Experience with MLOps best practices, model versioning, monitoring, and retraining workflows.
• Exposure to hospitality, property operations, guest experience, or enterprise safety data.
• Proven ability to translate complex analytics into practical business workflows and measurable impact.

Education

• Master's degree in computer science, statistics, data science, industrial engineering, operations research, mathematics, or related field preferred.
• Bachelor's degree with strong relevant experience acceptable.

Experience

• 5+ years of experience in data science, machine learning, risk analytics, or related area preferred.