Software Engineer, DICOM infrastructure

New Lantern is building an AI radiology resident—think Cursor for radiologists. Our application integrates the three software tools radiologists use all day [1], together with AI, to create a platform that saves radiologists toil and lets them focus on reading images and finding disease.

About This Role

You'll take ownership of New Lantern's existing OSS-based DICOM infrastructure, and help to operate and maintain it, all while planning, designing, and eventually implementing our next generation of DICOM infrastructure that leverages cloud services and architectural practices to improve reliability and our ability to scale, while reducing operational overhead.

What You’ll Do

• Own New Lantern’s DICOM image ingestion, routing, and storage infrastructure.
• Become an expert in operating our existing DICOM systems while designing the architecture for the next order of magnitude in scale and customer requirements.
• Design and build reliable, observable processing pipelines for images and binary data at scale.
• Monitor and optimize our DICOM infrastructure for throughput, latency, correctness, and cost.
• Support New Lantern’s integration team in establishing connectivity with customer PACS systems.
• Collaborate with radiologists and PACS admins/IT staff to discover requirements, build prototypes, and validate solutions.
• Partner with product and viewer teams to enable new products and workflows.

You have

• 6+ years backend/infrastructure experience, including building and owning at least one large‑scale data pipeline.
• Deep experience with a modern, object-oriented programming language, including a web framework.
• Systems programming & ops/admin experience (networking, linux/POSIX, concurrency, filesystems, memory management, etc.)
• Experience architecting and building systems on a major cloud platform.
• An end-to-end ownership mindset across design, implementation, testing, and monitoring.

Nice to haves

• Deep Python experience, including a Python web framework (ideally Django).
• Prior contributions to dcm4che(e)/orthanc/etc. or other image processing or medical imaging software.
• Knowledge of DICOM, PACS systems, and networking (addressing, routing, TLS, VPNs), both standards and industry practices.
• Experience with streaming and batch data pipelines and object storage at scale.

[1] a radiology image archive & viewer (PACS), a medical report dictation system, and work management system