Senior Data Scientist - Machine Learning

Accountabilities:

• • Design, train, validate, and refine supervised machine learning models that identify providers and billing patterns associated with healthcare fraud, waste, and abuse risk.
  • Develop robust labeling and entity-resolution approaches using investigative case data, payer feedback, public enforcement records, and exclusion data.
  • Design validation strategies for complex real-world conditions, including delayed labels, incomplete historical coverage, extreme class imbalance, data leakage risks, and rapidly evolving fraud schemes.
  • Engineer features directly against very large claims datasets using Python and SQL, collaborating with data engineering and business intelligence teams to ensure scalable processing.
  • Produce actionable model outputs that provide investigators with ranked risk scores, human-readable rationales, relevant claims, and supporting evidence.
  • Deploy machine learning models into production and establish reliable practices for scheduling, model versioning, monitoring, drift detection, and ongoing model health.
  • Help define and establish the modeling and deployment standards that will guide future data science initiatives.
  • Partner with fraud, waste, and abuse subject matter experts to distinguish genuine anomalies from patterns caused by coverage policies, claim edits, or other legitimate factors.
  • Communicate methodologies, assumptions, limitations, and analytical findings effectively to both technical and non-technical audiences, including clients and program stakeholders.
  • Contribute to a strong machine learning practice by establishing scalable approaches, documentation, and best practices for future development.

  Requirements:

  • Master’s degree in statistics, computer science, engineering, applied mathematics, or another quantitative discipline, or a bachelor’s degree combined with equivalent hands-on experience.
  • At least 5 years of experience building, validating, and delivering supervised machine learning models using real-world data, particularly where labels may be incomplete, delayed, or biased.
  • Demonstrated experience deploying machine learning models into production, including scheduling, version control, monitoring, and collaboration with data engineering teams.
  • Strong proficiency in Python and SQL, including experience performing feature engineering directly within large-scale data warehouses rather than relying on local data extraction.
  • At least 2 years of experience working with healthcare claims data, such as Medicare, Medicaid, or commercial claims, and familiarity with coding systems including ICD-10, CPT, HCPCS, and DRG.
  • Strong understanding of validation techniques for imbalanced and temporally shifting datasets, including out-of-time evaluation, leakage detection, calibration, and precision-focused ranking metrics.
  • Ability to translate complex model outputs into clear, understandable insights for investigators and other non-technical users.
  • Strong communication skills and the ability to explain and defend analytical methodologies to technical audiences while presenting findings effectively to clients and stakeholders.
  • Experience with graph or network analytics, entity resolution, and record linkage is highly desirable.
  • Knowledge of positive-unlabeled, semi-supervised, or active learning approaches is a plus, particularly for capacity-constrained investigative workflows.
  • Experience developing models in regulated or adverse-action environments where explainability, fairness, and responsible model use are important is beneficial.
  • Familiarity with anomaly detection, peer-group construction, case-mix methodologies such as HCC, AWS, Snowflake, Snowpark, or model lifecycle tooling is preferred.
  • Experience with healthcare fraud, waste and abuse analytics, program integrity, multi-payer datasets, coverage policies, or claims edits such as NCCI is an advantage.
  • Must be able to obtain and maintain a Public Trust.
  • Candidates must reside in the United States; work visa sponsorship is not available for this position.

  Benefits:

  • Salary: Expected range of $123,250–$166,750 per year, with actual compensation determined by experience, geographic location, and applicable contractual requirements.
  • Remote work: Fully remote position with a standard 40-hour workweek.
  • Healthcare: Multiple medical plan options, including plans with Health Savings Accounts, as well as dental and vision coverage.
  • Retirement: 401(k) plan with company matching and pre- and post-tax contribution options, subject to applicable IRS limits.
  • Paid time off: Vacation, sick, and personal leave, plus paid holidays. New employees typically receive 15 days of paid leave and 10 paid holidays annually, prorated according to hire date.
  • Family and personal leave: Paid parental, military, bereavement, and jury-duty leave, along with up to 160 hours of paid family leave during a rolling 12-month period for eligible employees.
  • Financial protection: Short- and long-term disability, life insurance, accidental death and dismemberment, personal accident, critical illness, and business travel and accident insurance.
  • Flexibility: Full-flex work arrangements designed to support work-life balance.
  • Travel: Less than 10% travel expected.
  • Career development: Opportunity to establish a machine learning practice, define production standards, and grow alongside professionals working on complex data science and engineering challenges.