Senior Data and Integrations Engineer

About Bluestone Health

Bluestone Health is building a more connected healthcare experience for employer groups and communities, including people in rural and remote areas where access to coordinated care is often limited. Our model brings together health plan administration, pharmacy benefits, primary care, specialty care, and care coordination so members are not left to navigate a fragmented system on their own.

We use software, data, and responsible AI to connect systems, reduce administrative work, and help care teams act on a more complete understanding of the member. Our team is small and high-trust. People here take ownership, communicate directly, and follow work through to a confirmed result.

Why Now

Bluestone needs a dependable data foundation for both daily operations and healthcare analytics. Eligibility, medical and pharmacy claims, clinical information, and operational records arrive from different organizations in inconsistent formats. We need to turn those inputs into reliable production flows and documented datasets that can support recurring reporting and useful analysis.

This is durable engineering work, not a role tied only to a client conversion deadline. The person who joins will help establish the data architecture, quality controls, and analytical foundation rather than inherit a finished platform.

The Role

We are looking for a senior, hands-on data engineer who can build and operate production pipelines and partner closely with our Healthcare Analytics and Client Reporting Lead.

You will own how data is ingested, transformed, validated, modeled, and made available. The Analytics Lead owns reporting questions, healthcare methodology, interpretation, and client-facing conclusions. You will help translate those requirements into tested TypeScript/Node.js and SQL workflows, with Python where appropriate, reusable datasets, and reliable analytical processes.

This is an engineering-first role with meaningful analytical depth. Applied data-science experience is valuable, but the core job is building dependable systems and making analyses repeatable. It is not primarily a machine-learning research position or a dashboard-only role.

A separate EDI & Implementation Lead owns the client-facing onboarding process. You will participate in technical discovery and decisions when needed, without being expected to run every vendor meeting or chase every implementation action item.

What You'll Own

• Production data pipelines: Build ingestion, transformation, validation, and delivery workflows for eligibility, enrollment, claims, pharmacy, clinical, and operational data using APIs, SFTP, scheduled files, queues, and event-driven processing.
• Analytical data models: Create documented warehouse tables and data marts with clear grain, keys, history, lineage, and source semantics. Handle changing eligibility, corrections, reversals, and late-arriving data explicitly.
• Technical partnership with analytics: Implement agreed cohorts, measures, and reporting datasets with the Analytics Lead. Build reusable TypeScript/Node.js and SQL workflows, with Python where appropriate, rather than a collection of one-off scripts.
• Data quality and reconciliation: Automate checks for missing files, duplicates, required fields, unexpected eligibility changes, inconsistent identifiers, and source-to-target differences. Distinguish source problems from transformation or analytical-logic problems.
• Reliability and recovery: Design idempotency, retries, safe backfills, controlled reprocessing, monitoring, alerts, and practical incident runbooks. Own technical root-cause correction for the systems you operate.
• Identity and context: Preserve client, group, plan, member, effective-date, and other meaningful distinctions across sources. Surface ambiguous rules rather than silently inventing mappings.
• Engineering standards: Use version control, automated tests, code review, infrastructure as code, and repeatable deployment practices. Make consequential transformations inspectable and auditable.
• Operational enablement: Give operations staff useful exception views and approved procedures. Automate recurring work where worthwhile without hiding errors or removing necessary review.
• Applied analysis: Support exploratory analysis, statistical validation, and practical models when the use case warrants them. Validate assumptions and results with the Analytics Lead and appropriate domain owners.

Ownership Boundaries

You own pipeline architecture, production code, technical mappings and transformations, warehouse models, and technical reliability. The EDI & Implementation Lead coordinates external requirements and acceptance; authorized business owners approve benefit and eligibility rules. Operations runs defined checks and gathers exception evidence. The Analytics Lead owns measure definitions, methodology, interpretation, and reporting narrative.

You should be able to communicate directly with partners on technical questions. You are not the primary account manager, the owner of every onboarding meeting, or the substitute for clinical, actuarial, or business judgment.

Current Environment

Our environment includes AWS, PostgreSQL, TypeScript/Node.js services, SST/infrastructure as code, APIs, SFTP, flat files, queues, and event-driven processing. Snowflake has been used for analytics. Node.js, TypeScript/JavaScript, and SQL are core to our current production work. We do not currently operate an established Python runtime; we expect to add Python execution environments where justified by data processing or analytical needs.

You do not need experience with every service. You do need practical cloud and data fundamentals and the ability to work within an existing AWS-first engineering stack rather than introduce unnecessary platform complexity.

What We're Looking For

Required

• Senior-level experience building and operating production data pipelines, backend data services, or analytical platforms.

• Excellent SQL, including complex joins, incremental processing, reconciliation, performance investigation, and maintainable data modeling.

• Strong production experience with Node.js and TypeScript/JavaScript, including building, testing, debugging, and operating backend services and data-processing workflows.

• Production experience with AWS or another major cloud platform and integrations using APIs, files, object storage, or queues.

• Experience with testing, version control, deployment, monitoring, failure recovery, and documented data lineage.

• Ability to translate analytical requirements into dependable datasets and explain discrepancies across source, transformation, and reporting layers.

• Clear communication, independent investigation, and end-to-end ownership in a small team.

• U.S. residence and substantial overlap with the U.S. team.

Strongly Preferred

• Healthcare eligibility, medical/pharmacy claims, payer, TPA, PBM, or EHR data experience.
• Familiarity with member-month denominators, claim reversals and adjustments, runout, effective dates, and healthcare coding.
• Snowflake or comparable warehouse experience; tested transformation models and governed analytical datasets.
• Practical statistical judgment about cohorts, selection bias, uncertainty, and the limits of causal conclusions.
• Hands-on Python for data processing, validation, and analytical workflows; experience deploying and operating Python workloads is particularly useful as we add this capability.
• AWS serverless systems, PostgreSQL, and infrastructure as code.
• HIPAA, minimum-necessary access, auditability, and security controls for sensitive data.

Nice to Have

• X12 experience, particularly 834 eligibility/enrollment files.
• Applied forecasting, risk modeling, or other data-science work delivered into maintainable workflows.
• Experience improving a small company's data platform without unnecessary operational overhead.

What Good Looks Like

Important feeds are monitored and recoverable. Analysts can use documented datasets without repeatedly reconstructing source meaning. A recurring client report can be rerun from governed inputs with tested calculations, reconciled totals, and traceable assumptions. Manual checks become smaller and more focused, and the CTO is no longer the default debugger or translator.

Security and Responsible AI

This role works with systems and datasets containing protected health information. Follow Bluestone's access, security, privacy, and HIPAA policies; use approved devices and systems; and preserve auditability. Sensitive data must not enter unapproved AI tools. AI may accelerate development and documentation, but you remain accountable for correctness, security, and maintainability.

Expected Hiring Process

1. Application or recruiter introduction.
2. Screening conversation covering production data ownership, analytical collaboration, and working hours.
3. Paid, time-bounded practical exercise using synthetic data: ingestion/transformation, reconciliation, analytical modeling, and recovery considerations, with Node.js/TypeScript and SQL as the core implementation languages.
4. Technical working session reviewing the exercise, assumptions, tests, architecture, and tradeoffs.
5. Final conversation on mutual fit, logistics, and offer details.

We aim to complete the process in approximately two weeks once interviews begin.

How to Apply

Send your resume and a brief note describing a production data system you owned, how you made its outputs trustworthy, how you supported analysts or domain experts, and your location, time zone, and availability.

We review applications personally. We do not use automated keyword screening to decide who gets considered.

Equal Opportunity

Bluestone Health is committed to providing equal employment opportunities without discrimination or harassment on the basis of race, color, religion, sex, gender, sexual orientation, gender identity, national origin, disability, age, veteran status, or any other status protected by applicable law.