SmartLight AnalyticsRemote
Healthcare Statistical Data Scientist
United States of America
Full Time
9 hours ago
H1B Sponsor Likely
Key skills
Pythonscikit-learnXGBoostTensorFlowPyTorchMachine learningHealthcare claims datasetsCPT/HCPCSICD-10DRGNDCStatistical modelingData mining techniquesSQLMachine LearningMLDeep LearningNLPMLOpsAnalyticsData MiningCI/CD
About this role
SmartLight Analytics is dedicated to improving the healthcare system by combating fraud, waste, and abuse through advanced data analysis. They are seeking a Healthcare Statistical Data Scientist to design, build, and deploy machine learning models that transform complex healthcare datasets into actionable insights for improving cost efficiency and care quality.
Responsibilities:
- Develop, train, and deploy ML models for use cases such as:
- Claims cost prediction and utilization forecasting
- Fraud, waste, and abuse detection
- Risk adjustment and member stratification
- Provider performance and network optimization
- Apply modern ML techniques including gradient boosting, deep learning, NLP, and probabilistic modeling
- Capable of applying advanced predictive analytics to correlate disparate datasets and events and derive business value
- Build scalable pipelines for feature engineering, model training, validation, and monitoring
- Analyze and interpret medical, pharmacy, and dental claims (CPT/HCPCS, ICD‑10, DRG, NDC)
- Translate domain knowledge into meaningful features and model strategies
- Partner with clinicians, product managers, and business stakeholders to define problems and measure outcomes
- Communicate complex analytical findings in clear, actionable terms
Requirements:
- Strong proficiency in Python and ML libraries (scikit‑learn, XGBoost, TensorFlow/PyTorch)
- Hands‑on experience with healthcare claims datasets and coding systems
- Solid understanding of statistical modeling, machine learning algorithms, and data mining techniques
- Strong knowledge and expertise working with SQL
- Ability to translate business needs into analytical solutions
- Must have demonstrated the ability to solve complex problems with minimal direction
- Experience with NLP applied to clinical notes or unstructured healthcare data
- Familiarity with actuarial concepts, risk scoring, or value‑based care models
- Familiarity deploying models into production (MLOps, CI/CD)
- Background in health economics, epidemiology, or biostatistics
- Prior work with FHIR, HL7, or interoperability standards