Develop and deploy fraud detection models for LTC insurance claims using statistical and machine learning techniques
Perform exploratory data analysis (EDA), feature engineering, and hypothesis testing to identify fraud patterns and anomalies
Build, evaluate, and optimize traditional statistical models as well as tree‑based models such as Random Forest, XGBoost, CatBoost, LightGBM etc.
Independently conduct data analysis, research, and model experimentation, and translate findings into actionable insights
Write clean, efficient, and production‑ready code using Python and SQL
Work extensively with large datasets using cloud platforms, primarily Google Cloud Platform (GCP)
Query and manage data using BigQuery, and handle datasets stored in Cloud Storage (Buckets)
Use Git for version control, collaboration, and code review
Prepare clear, concise, and impactful presentations for clients, explaining analytical findings to both technical and non‑technical stakeholders
Collaborate with business, data engineering, and client teams to ensure models align with fraud investigation and business objectives

6–7 years of hands‑on experience in data science, analytics, or applied machine learning
Strong understanding of statistical modeling, probability concepts and hypothesis testing
Proven experience with tree‑based and ensemble machine learning models (RF, XGBoost, CatBoost, LightGBM)
Expert‑level SQL for data extraction, transformation, and analysis
Strong Python skills for data analysis and modeling
Experience using Git for source code management
Solid exposure to cloud‑based analytics environments, preferably Google Cloud Platform (GCP), BigQuery and Cloud Storage
Ability to work independently, manage deliverables, and drive tasks end‑to‑end
Excellent verbal and written communication skills, essential for a client‑facing role

Data Scientist

Key skills