ML Research Engineer / Scientist (Remote, international)

Role overview A research-engineering role developing foundation models that interpret full volumetric CT studies rather than single slices. The work targets clinical deployment: trained models move through regulatory clearance into live hospital reading workflows, so research has a short path to patients. The team pairs ML engineers with fellowship-trained radiologists across multiple subspecialties who read alongside the model every day.

Responsibilities - Train and improve foundation models that consume an entire CT study end to end - Develop vision-language learning that ties image features to the language radiologists actually write - Build models that flag multiple urgent findings in one pass, calibrated for radiologist oversight - Run distributed training and large-scale evaluation pipelines - Drive experiments from first idea through the operating point that ships

Requirements - Deep PyTorch fluency, including custom architectures, training loops, and distributed training - Working understanding of why a loss or architecture works, beyond how to call it - Comfort designing and running your own experiments without hand-holding - Commitment to rigor and reproducibility given the clinical weight of the work

Benefits and work setup - Fully remote, open to candidates worldwide - Compensation set to the local market for the country where the work is done; specifics shared early in the process - International offers are cash-only, with the trade-off explained up front