Scientific Data Engineer

Responsibilities

• Design, implement, and maintain reusable Python libraries for scientific and image analysis and quality control.
• Build and operate reproducible analysis pipelines integrated with workflow-management and scientific data systems.
• Define data models for raw data, experimental metadata, derived results, and analysis provenance across experimental assays.
• Partner with wet-lab and computational scientists to translate scientific questions into requirements, validated methods, and maintainable software.
• Establish testing, validation, versioning, observability, and failure-handling practices for scientific workflows.
• Evaluate external methods and tools, make build-versus-buy decisions, and define evolvable system interfaces.
• Balance immediate experimental needs with shared infrastructure investments that improve consistency, interoperability, and research velocity.

Requirements

• Strong software engineering experience in Python, including API design, testing, packaging, code review, and collaborative development with Git.
• Experience designing and implementing image-analysis pipelines for scientific data and selecting, configuring, and validating classical or learned methods.
• Experience with rigorous scientific computing practices, including quantitative validation, quality control, reproducibility, provenance, and explicit failure-mode handling.
• Experience building production or research workflows with an orchestration system such as Dagster, Prefect, or Airflow.
• Experience working directly with wet-lab scientists to clarify ambiguous needs, communicate trade-offs, and develop scientifically valid solutions.
• Ability to distinguish assay-specific requirements from reusable infrastructure and make trade-offs involving delivery speed, maintainability, reliability, performance, and cost.
• Demonstrated ownership and judgment in breaking ambiguous work into milestones, surfacing risks, and independently driving projects to completion.
• Helpful qualifications include experience with Bazel or another large monorepo build system, next-generation sequencing, spatial omics, ultrasound, other scientific data modalities, or early-stage research environments.