Lead & Senior Data Engineer

Lead & Senior Data Engineers

Location: Remote United States (US work authorization required)

Job Type: Multi-month engagement, likely to extend

Openings: 4 positions, 2 Lead and 2 Senior

Start Date: Around December 1, 2026

About Virtasant

Virtasant is a global technology services company with a network of over 4,000 technology professionals across 130+ countries. We specialize in cloud architecture, infrastructure, migration, and optimization, helping enterprises scale efficiently while maintaining cost control.

Our clients range from Fortune 500 companies to fast-growing startups, relying on us to build high-performance infrastructure, optimize cloud environments, and enable continuous delivery at scale.

About the Role

We are building a team for a large enterprise program in healthcare operations and workforce decisioning. The platform pulls data from the many systems a hospital runs on, maps it into a shared ontology and knowledge graph, and uses it to help executives make operational decisions, act on them and measure the results.

The hardest recurring problem is getting messy enterprise data in, clean and trusted, for every new customer. You will build that: connectors and pipelines from HR, scheduling, EHR and external benchmark sources into Snowflake on Azure, mapped to the ontology and published to a Neo4j graph. You will make it repeatable, so the next customer’s data goes live in days, not months. You will work with real, sensitive client data, and you will use AI-assisted development as a normal part of how you build.

Lead vs Senior

Both levels are hands-on and write production code every week.

Lead: Owns the technical direction of a pod: its design, its standards and the order in which work gets done. Works directly with the client’s architects and data owners to agree source extracts, data contracts and acceptance criteria. Reviews other engineers’ code and designs, and is accountable for what the pod ships. You have led a small team of engineers, set technical direction and worked directly with client architects and stakeholders.

Senior: Owns significant pieces end to end, such as a source connector, a set of mappings or the reconciliation for a domain, inside the direction the Lead sets. Works independently, raises problems early and needs no day-to-day supervision.

What We're Looking For

• You have integrated messy enterprise source systems in production, not just clean APIs. You can talk through two or three: what the source really looked like, how you handled late files, schema drift, duplicates and conflicting records, and what you would do differently.

• Strong SQL and Python, and hands-on Snowflake. You have built and run layered pipelines (raw, cleaned, modelled) with dbt or equivalent, with mappings and rules kept as versioned code.

• Data contracts and versioned schemas. You define what a source must send, validate it on arrival, quarantine what fails and evolve schemas without breaking what sits downstream.

• Data quality, lineage and reconciliation as part of the build. You can prove that totals match the source, trace any number back to the record it came from, and rerun a load without duplicating data.

• Regulated or sensitive data. You have worked under PHI/PII controls such as HIPAA: masking and tokenisation, least-privilege access, audit trails and keeping sensitive fields out of places they should not be.

• Multi-tenant data platforms with strict per-customer isolation. You have built onboarding that is repeatable: a new customer is configuration and mapping, not a new code base.

• Azure experience: storage, identity and infrastructure as code (Terraform or Bicep).

• AI-assisted development. You use coding assistants daily and check what they produce.

• You must be based in the United States and authorized to work in the US. Occasional domestic travel may be required.

• Available around December 1, 2026.

Strong Plus

• Healthcare data: HR and workforce systems such as Workday or UKG, scheduling and timekeeping, or EHRs such as Epic, Cerner or MEDITECH.

• Graph databases, ideally Neo4j: loading a graph from a warehouse and modelling for it.

• Mapping data into an ontology or a common data model shared across many customers.

• Entity resolution: matching people and records across systems without a shared identifier.

• External benchmark or reference data feeds, by data share or flat file.

• Batch pipelines built to be event-ready, with idempotent loads and incremental refresh.

• Open lineage and metadata standards such as OpenLineage.

• Monitoring data feeds in production: freshness, volume and failure alerts.