Imaging Data Engineer

This is a highly hands-on role for someone who enjoys solving difficult data problems. You will work directly with massive collections of DICOM and other medical imaging files, develop automated data pipelines, troubleshoot imaging and metadata issues, and help transform raw clinical imaging into high-quality datasets that are ready for AI development.

What You’ll Do

• Build, maintain, and improve pipelines for ingesting, processing, organizing, validating, and exporting large medical imaging datasets.
• Work extensively with DICOM files, including DICOM metadata, pixel data, medical reports, demographic data, and other patient/study information.
• Develop automated workflows for de-identification and quality control of medical images and associated metadata.
• Process and manage very large datasets containing millions of imaging studies and potentially terabytes or petabytes of data.
• Utilize tools to identify, extract, transform, validate, and organize imaging data from complex folder structures and disparate source systems.
• Automate repetitive data-management tasks using Python, PowerShell, SQL, APIs, and other scripting/engineering tools.
• Develop processes to identify missing, corrupt, duplicated, inconsistent, or otherwise problematic imaging data.
• Analyze and manipulate large datasets containing patient, study, series, accession, modality, demographic, and clinical metadata.
• Work with cloud and object storage environments such as AWS S3, Google Cloud Storage, Azure, Backblaze B2, and similar platforms.
• Work with imaging transfer technologies including DICOM, DICOMweb, PACS, cloud storage, secure file transfer, and large-scale data migration.
• Build processes for dataset curation, including cohort selection, deduplication, date shifting, metadata normalization, and quality assurance.
• Help develop internal tools that make it easier for our team to search, inspect, validate, and work with massive imaging datasets.
• Investigate unusual imaging/data problems and develop practical solutions rather than simply escalating them.
• Collaborate with data scientists, AI developers, imaging experts, and healthcare organizations to understand and solve data requirements.
• Help establish repeatable, documented, and auditable processes for producing high-quality medical imaging datasets.

We’re looking for someone who is technically strong, curious, resourceful, and comfortable taking on challenging data projects.

Required

• Strong programming/scripting experience in Python, PowerShell, or a similar language.
• Strong understanding of filesystems, databases, data structures, and data-processing workflows.
• Experience working with large datasets and automating data-intensive processes.
• Strong Excel/CSV/data manipulation skills.
• Ability to troubleshoot problems independently and develop practical solutions.
• Strong attention to detail. In medical imaging data, a single incorrect metadata field can matter.
• Ability to work comfortably from the command line and automate tasks rather than performing repetitive work manually.
• Strong organizational and documentation skills.

Highly Preferred

Experience with one or more of the following:

• DICOM / medical imaging
• Certified Imaging Informatics Profession (CIIP) certification
• PACS or medical imaging systems
• Radiology, cardiology, pathology, dental, or other clinical imaging data
• Medical image de-identification
• Python imaging/data libraries
• SQL
• AWS, Google Cloud, Azure, or other cloud platforms
• Object storage such as S3 or GCS
• Docker or other container technologies
• REST APIs
• PowerShell
• Linux
• ETL/ELT pipelines
• Large-scale data migration
• Data quality and validation frameworks
• Machine learning or AI datasets, You may be a particularly strong fit if you:
• Enjoy working with complex datasets.
• Like understanding how things work at a low level.
• Enjoy automation and eliminating repetitive manual processes.
• Are comfortable learning technologies that you haven’t used before.
• Can move between Python, PowerShell, SQL, command-line tools, and cloud environments as needed.
• Have an interest in medical imaging, healthcare, or AI.
• Are equally comfortable writing code and troubleshooting a strange data problem.
• Care about data quality and understand that “mostly correct” isn’t good enough for healthcare data.
• Enjoy working in a small, highly entrepreneurial environment where each team contributes in a variety of ways.

At Vega Imaging Informatics, you’ll have the opportunity to work on problems at the intersection of healthcare, medical imaging, and artificial intelligence.

Our work helps make high-quality medical imaging data available for organizations developing the next generation of AI technologies.

You’ll have significant exposure to large-scale imaging infrastructure, DICOM, de-identification, cloud storage, dataset engineering, and the practical challenges of building AI-ready healthcare datasets.

This is an opportunity to have a direct impact on the infrastructure behind medical imaging AI, rather than working on a small piece of a large corporate system.

Compensation & Benefits

Competitive compensation based on experience and technical expertise.

Vega offers a generous benefits package including medical, dental, vision, paid holidays, 401(k) matching, paid time off, flexible work hours, computer equipment provided, cell phone reimbursement, professional development funding., * 401(k) matching

• Dental insurance
• Flexible schedule
• Health insurance
• Paid time off
• Professional development assistance
• Vision insurance

Vega Imaging Informatics is looking for an exceptional Imaging Data Engineer to help us build and manage some of the world’s largest and most diverse collections of de-identified medical imaging data., You’ll work with real medical imaging data at very large scale and help AI Developers design solutions to improve healthcare around the world.

One day you might be developing a pipeline to process hundreds of thousands of DICOM studies. The next day you might be investigating why a particular vendor’s images contain unexpected metadata, building a script to reconcile millions of records, or figuring out how to efficiently move terabytes of imaging data between storage environments.

We value engineers who don’t just say “that’s not possible.”

We want people who say