Principal Data Engineer

Vesalius Therapeutics•Cambridge, MA

1d•$179,000 - $236,000

About The Position

Vesalius Therapeutics is seeking a Principal Data Engineer to lead the architecture and evolution of the data platform supporting its human genetics, multi-omics, and high-content imaging programs. This senior, hands-on leadership role has broad technical authority and serves as the architectural lead for data engineering, setting direction for the AWS environment and the pipelines, catalog, and access layers built on it. The role defines the technical roadmap and standards for the platform. The work spans cloud data engineering—including orchestration, warehousing, infrastructure as code, and cost governance—and scientific data challenges such as connecting microscopy images to their cell lines, donors, and studies, and enabling scientists to answer genotype questions directly. The platform is established and productive and is entering a deliberate modernization phase. The Principal Data Engineer will lead this transition and substantially influence the next-generation architecture, while supporting a rapidly expanding research organization and partnering with machine learning and computational biology leaders.

Requirements

10+ years building and operating production data infrastructure, with a track record of architecting and owning platforms end to end.
Deep AWS expertise, including S3, IAM, EC2, Batch, Lambda, and Spark (EMR or equivalent).
Experience as technical owner of production cloud environments, including multi-account architecture and cross-account access.
Strong Python and SQL skills, with a background in production-grade pipeline development.
Production experience with workflow orchestration such as Nextflow, Airflow, Temporal, Step Functions, or Redun, and the judgment to evaluate and select among tools.
Experience with infrastructure as code (CloudFormation/CDK), CI/CD, containers, and observability; DevOps is a core competency.
Data modelling and warehouse or lakehouse design experience, including metadata management, lineage, and data quality practice.
Cost governance at scale, including budgets, tagging, rightsizing, and storage lifecycle across hundreds of terabytes.
Fluency with scientific data semantics, including studies, samples, donors, cell lines, and assays, and ability to partner directly with scientists to translate questions into data models.

Nice To Haves

Familiarity with genetics and genomics data.
Experience with microscopy or high-content screening data, including large binary object stores, transfer optimization, and image streaming for model training.
Production experience with Databricks, OpenMetadata, or a comparable catalog.
Snowflake experience, particularly governed access patterns over cloud object storage.
ML infrastructure experience with SageMaker or Databricks ML, MLflow or Weights & Biases, and feature stores.
Experience serving data to LLM or agent clients through MCP servers or tool APIs.
Experience leading data platform architecture as a senior or principal individual contributor, ideally in biotech, pharma R&D, or a research institute.

Responsibilities

Serve as technical owner and architect of the AWS environment, including account structure, IAM, networking, compute, and storage.
Lead and extend cost governance, including budgets, alerting, tagging enforcement, rightsizing, and storage lifecycle and tiering.
Define and own developer operations for the data platform, including infrastructure as code, CI/CD, container builds, and secrets and environment management.
Upgrade the data lake, managed orchestration, metadata, and data ingestion workflows to scale with future data production.
Enhance the infrastructure that transfers data from the lab to the cloud.
Partner with machine learning and computational biology leaders to define the interface between the platform and modelling.
Lead the modernization of the established data platform and influence its next-generation architecture.
Connect microscopy images to their cell line, donor, and study of origin, and enable scientists to answer genotype questions directly without specialist intervention.