Senior Machine Learning Engineer – Clinical Team

Midjourney

. Own tissue-class segmentation and labeling models for the ultrasound CT clinical analysis layer and the pipelines that make them trainable and verifiable .

Posted 9/18/2026full-timeCalifornia • United StatesSeniorWebsite

Core Competencies

Role fit

Core Competencies

Use this summary to align your resume positioning with the role.

Demonstrates expertise in developing and evaluating machine learning models for medical imaging, particularly in tissue-class segmentation and labeling. Proficient in deploying models within HIPAA-compliant environments while ensuring quality control and reproducibility.

Highest-signal resume keywords

Applied ML ExperienceImage SegmentationDeep-Learning Tooling (PyTorch)U-Net/nnU-Net ExperienceML Data Curation

ATS Keywords

Tailor your resume

Applicant Tracking System Keywords

Tip: use these terms in your resume and cover letter to boost ATS matches.

Hard Skills

Model ArchitectingModel TrainingModel EvaluationSemantic SegmentationInstance Segmentation3D Volumetric SegmentationSelf-Supervised LearningMasked AutoencodersContrastive PretrainingSimulation-Driven Pretraining

Soft Skills

CollaborationQuality ControlDocumentation

Tools & Technologies

MONAIITK/SimpleITK3D SlicerDINOSimCLR

Industry Keywords

Medical ImagingAnatomyBody CompositionHIPAA ComplianceCT ImagingMRI ImagingUltrasound Imaging

Tech Stack

Tools & technologies

CloudPyTorch

About the role

Key responsibilities & impact

Own tissue-class segmentation and labeling models for the ultrasound CT clinical analysis layer and the pipelines that make them trainable and verifiable
Retune models across 2D per-slice, 3D volumetric, and 2D×3D fusion as reconstructed image inputs and clinical indications evolve
Define training and evaluation pipelines, datasets, and metrics; map model behavior to user needs and design requirements
Work with data labeling contractors, expert clinicians, and internal cloud/data teams on labeling specifications, quality control, and dataset versioning
Productionize models into a versioned, HIPAA-bound analysis service with reproducible, low-latency inference, per-prediction confidence, drift monitoring, and safe fallbacks

Requirements

What you’ll need

Strong applied ML experience developing new models, including architecting, training, and evaluating from scratch and benchmarking existing models
Experience with image segmentation, including semantic/instance segmentation in 2D and ideally 3D/volumetric formats
Ability to work between open-ended research iteration and quantifiable, testable models
Fluency in modern deep-learning tooling such as PyTorch and current development practices
Ability to work under design controls with documentation and verification requirements
Experience with U-Net/nnU-Net, 3D U-Net, transformer-based and promptable segmentation such as SAM
Knowledge of voxel- and mesh-level prediction geometry and coordinate frames
Experience with self-supervised or semi-supervised learning, masked autoencoders, contrastive pretraining such as DINO/SimCLR, active learning, weak labels, or simulation-driven pretraining
Hands-on ML data curation, ground-truth definition, and labeling or simulation pipelines using MONAI, ITK/SimpleITK, or 3D Slicer
Experience with segmentation models for ultrasound imaging
Experience with ML for imaging or inverse problems in CT, MRI, ultrasound, or adjacent physics-based domains
Experience deploying models in versioned, auditable, high-stakes settings
Background in anatomy, medical imaging, or body composition is a plus