Contract Health Data Analyst

TLDR:_ Build our data warehouse schemas in dbt on AWS Glue. Health insurance experience required We’re upgrading our plan intelligence product from great to world class, and we need someone who knows health insurance data cold. You’ll design the schemas and models that power our data warehouse using dbt and AWS Glue. Your work will directly support how we analyze plan performance and generate recommendations for employers, providers, and TPAs. You’ll work directly with the CEO and partner across product and engineering. You’ll move fast, work on a unique dataset, and help shape a system that drives better health and lower costs for millions of Americans. What you’ll do Build and maintain our data warehouse schemas in dbt , running on AWS Glue . Model 50+ health plan and claims tables into clean, reliable, analytics-ready structures. Create fast, scalable SQL models that power dashboards, benchmarks, and plan recommendations. Develop frameworks for utilization analysis, risk stratification, pricing insights, and cost drivers. Work with sensitive claims data and ensure quality, accuracy, observability, and compliance. Iterate quickly with AI tools to ship analysis and data models faster than a traditional workflow. What you must have 2+ years of data analysis or data science experience in health care or insurance . Deep experience with claims, eligibility, pricing logic, or actuarial-style data . Strong SQL (complex joins, window functions, performance tuning). Experience with dbt, ETL pipelines, and cloud data stacks. AWS familiarity, ideally S3, Glue, Lambda, Redshift, or Athena. Ability to understand messy business questions and turn them into clean models. Comfort using AI tools like Claude or Cursor to boost productivity. Strong pluses Experience at companies like Flatiron, Komodo, Oscar, or health tech with claims-heavy datasets. Understanding of CPT, ICD-10, HCPCS, EDI 837/835, payer logic, and cost sharing models. Familiarity with data quality frameworks like Great Expectations. Exposure to ML workflows for risk or utilization models.