Data Analytics Engineer

Data Analytics Engineer Why This Role Exists We are building a semantic layer that makes our healthcare data platform queryable through structured, trustworthy metric definitions. We need someone who can own the metric build pipeline: define what a metric means, write the SQL that produces it, validate it against real claims data, and document it clearly enough that a machine or a human can trust the result. This is not a modeling or ML role. The work is precise, methodical, and high-leverage: you are building the measurement layer that everything else depends on. If you like making data reliable and correct more than making it flashy, this is your role. What You’ll Do

• ·Maintain and extend existing data mart builds: claims aggregation, financial metrics, clinical metrics, and curated analytical tables that power our SaaS platform
• · Author, test, and maintain metric SQL templates against our canonical data model (medical claims, pharmacy claims, eligibility, provider, plan design)
• · Build and maintain data pipelines in Python and SQL (Redshift, PostgreSQL, S3-based integrations)
• · Profile and validate third-party data feeds during vendor onboarding (new data sources land regularly)
• · Investigate data quality issues from warehouse transformation logic through final output
• · Produce analytical deliverables for clients and internal stakeholders (PMPM trending, cohort analysis, cost drivers)
• · Write clear documentation for metric definitions, data lineage, assumptions, and known limitations
• · Participate in monthly rotation for support triage (incoming tickets), release management, and mart build monitoring
• · Participate in code reviews and contribute to team standards, To be direct: if you are looking for a role focused on machine learning, deep learning, or model building, this is not it. We have ML work on the team, but this seat is about building the measurement and data infrastructure those models depend on. The work matters because nothing downstream can be better than the metrics feeding it. What Success Looks Like First 30 days: Understand the data model, mart structure, and toolchain. Complete onboarding. Make your first contribution to an existing mart build or pipeline (bug fix, new column, documentation). First 90 days: Own maintenance of one or more existing mart builds. Handle data profiling for at least one vendor integration independently. Begin authoring new metric definitions with guidance from senior team members. Be a reliable contributor in code reviews. First 6 months: Carry 3-5 active workstreams across mart maintenance and new metric development. Investigate and resolve data quality issues without hand-holding. Be the person the team trusts to take a business question and produce a clean, documented, testable metric, whether that lives in an existing mart or a new template.

• · 3-5 years in analytics engineering, data engineering, or a quantitative analyst role where you owned metric definitions
• · Strong SQL (window functions, CTEs, complex multi-table joins across large datasets; MPP warehouse experience preferred)
• · Python for data work (pandas, scripting, automation, testing; not just notebooks)
• · Experience building metrics from raw transactional data and validating them against business expectations
• · Comfort working across the stack: you’ll touch transformation logic, warehouse tables, and documentation
• · Clear written communication; you’ll document metrics for both engineers and business stakeholders

• · Git-based development workflow; pull requests and code review Strongly Preferred:

• · Healthcare analytics background: medical/pharmacy claims, eligibility data, employer-sponsored health plans, or adjacent domains (insurance, benefits, actuarial)
• · Familiarity with healthcare data concepts (PMPM, allowed amounts, service categories, plan design, enrollment periods)
• · Experience with SQL-based transformation frameworks or structured mart build processes

• · Exposure to data quality tooling or systematic data profiling Nice to Have:

• · Experience with AI-assisted development tools (Claude Code, Copilot, or similar)
• · Familiarity with Parquet, Iceberg, or lakehouse file formats
• · Statistical literacy (can interpret a regression, knows when a sample size is too small, understands variance)
• · Visualization experience (Tableau, Looker, or similar)