Data Engineer
About the role
Our product depends on bringing together data that has never previously lived in one system: insurance plans and networks, provider and pricing data, employer and member information, claims and medical records, and the activity generated across our own products and operations. We are looking for our first Data Engineer to build the foundation that makes this information reliable and usable across the company. This person will own the systems that ingest data from APIs, files, vendors, customer systems, public sources, and legacy portals; model it into durable representations; and make it available to our products, AI agents, operators, and business systems. This is a hands-on role with broad ownership across the data stack. You will make early architectural decisions, build the first versions yourself, and determine where rigor matters immediately and where simple systems can evolve. You will also help define how we use AI in data engineering to acquire, interpret, reconcile, and operate data more effectively. The right candidate has strong data engineering judgment, enjoys building from zero, and is excited to work with messy healthcare data where accuracy matters.
Responsibilities
- Build our data foundation: Own the systems that move, transform, organize, and serve data across the company, including the data warehouse (e.g. Databricks), ingestion and export infrastructure, orchestration and scheduling (e.g. Airflow or Dagster), transformation and modeling tools (e.g. dbt), semantic layers, data catalogs, quality, and observability.
- Unify fragmented healthcare data: Ingest and connect data from paid vendors, carriers, providers, government sources such as CMS, customer and benefits administration systems, public files, APIs, legacy portals, and our own applications.
- Build durable workflows for large files, changing schemas, incomplete records, and sources that were never designed to work together.
- Make the data trustworthy: Design canonical models for plans, networks, providers, employers, members, claims, and other core concepts. Preserve historical state, lineage, provenance, and source-specific nuance so our products, agents, operators, and customers can rely on the results.
- Put data into production: Make curated data available wherever it creates value, including our backend, member and employer experiences, AI agents, brokerage workflows, CRM, reporting, and other operational tools. Build reliable systems for moving data both into and out of the warehouse.
- Use AI to improve data engineering: Apply models and agents to acquire information from portals and older websites, interpret unfamiliar files, map changing schemas, reconcile conflicting sources, investigate quality issues, and automate work that has traditionally been manual.
- Set the roadmap and standards: Work across product, engineering, brokerage, and operations to identify the highest-leverage data capabilities, prioritize the roadmap, and decide where we need durable infrastructure now and where a simpler solution can evolve.