ML Research Engineer / Scientist (Boston, hybrid)

a2z Radiology AI•Boston, MA

48d•Hybrid

About The Position

Build the model that reads the whole scan. For over a decade, radiology AI has meant narrow tools that flag one finding and stop. We think that is a dead end. We are building the model that reads a CT the way a radiologist does: the whole study, drafted into the report. Your job is to make that model better, faster than anyone expects. This is research with a short line to patients. The models you train go through FDA clearance and deploy into live hospital reading rooms. The experiment you run this month can change how real studies get read next quarter.

Requirements

Proficiency in PyTorch, including custom architectures, training loops, and distributed training.
Deep understanding of why a loss or an architecture works, not just how to call it.
Ability to design and run your own experiments independently.
Commitment to rigor and reproducibility, as these models carry clinical weight.
Must be authorized to work in the United States.

Nice To Haves

Experience with medical imaging, 3D, or volumetric data.
Experience with vision-language models or self-supervised learning.
Background with DICOM, CT, or radiology.
A PhD or publications.

Responsibilities

Build foundation models that read an entire CT study, not one slice at a time.
Develop vision-language learning that ties what the model sees to the language radiologists actually write.
Create one model that catches many urgent findings at once, calibrated to operate safely in front of a radiologist.
Train and evaluate models at a scale most researchers never get to touch.
Conduct research that ships: your models reach FDA submissions, hospital deployments, and patients.
Own your models and experiments end to end, from first idea to the operating point that ships.