ML Research Engineer / Scientist (Remote, international)

Role overview

This is a research engineering role focused on building foundation models that interpret complete CT studies rather than flagging isolated findings. The position sits at the intersection of ML research and clinical impact, with trained models progressing through regulatory clearance and into hospital reading rooms on short timelines. It is fully remote and open to candidates worldwide.

Responsibilities

- Design and train foundation models that consume entire CT studies end to end, not single slices - Develop vision-language learning approaches that connect imaging features to the language clinicians actually write - Build unified detectors that surface multiple urgent findings simultaneously, calibrated to operate safely in front of a reviewing clinician - Own experiments from initial concept through the operating point that ships to deployment - Train and evaluate at large clinical scale, with models moving toward regulatory submissions and live hospital use within months - Collaborate closely with ML engineers, software engineers, and fellowship-trained clinicians across multiple specialties who read alongside the model

Requirements

- Deep fluency with PyTorch, including custom architectures, training loops, and distributed training - Strong grasp of why a chosen loss function or architecture works, not just how to call it - Track record of designing and running independent experiments without close direction - Commitment to rigor and reproducibility given the clinical weight of the models - Substantial experience building and shipping ML models end to end

Nice to have

- Background in medical imaging, 3D, or volumetric data - Familiarity with vision-language models or self-supervised learning - Experience with DICOM, CT, or radiology workflows - PhD or peer-reviewed publications - Prior medical-AI experience is not required; radiology will be learned from the clinicians on the team

Benefits and work setup

- Fully remote, open to candidates worldwide - Cash-weighted base salary set by the local market for the country where the work is done, with the specific range shared early in the process - International offers do not include equity - Small team environment where research has a short, direct line to patient impact