SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Rad AI is seeking a Senior ML Research Scientist to own multimodal machine learning work-streams in medical imaging and radiology. You will translate clinical and product needs into clear ML objectives, design and build modern ML systems including transformers, self-supervised learning, weak supervision, detection, localization, and segmentation. Your work will span image, report, and clinical data to develop systems that integrate seamlessly into real radiology workflows.
Key responsibilities include designing rigorous evaluations that go beyond aggregate metrics to include clinically meaningful operating points, robustness, calibration, and performance across relevant data slices. You'll partner with engineering teams to productionize models, make practical system tradeoffs, and learn from real-world performance post-launch. You'll investigate failure modes such as laterality errors, poor grounding, hallucination, dataset bias, domain shift, and workflow disruption.
You'll communicate research findings through design documents, experiment reviews, and presentations to both technical and clinical partners. You'll contribute to the research roadmap by identifying promising approaches and helping the team prioritize what to pursue. Mentoring less experienced researchers and engineers through collaboration, code reviews, and technical guidance is also part of the role.
Required qualifications include strong applied experience in computer vision, NLP, or deep learning with a track record of independently designing experiments and turning findings into working systems. You need deep hands-on ability in Python and PyTorch, with strong intuition for model architecture, data quality, experimentation, and evaluation. Experience with modern vision or multimodal techniques such as vision transformers, contrastive learning, or masked image modeling is essential. You should have 4+ years of relevant applied ML research or engineering experience (or equivalent scope and impact) and an MS, PhD, or equivalent in Computer Science, Electrical Engineering, Machine Learning, or a related quantitative field.
Nice-to-have qualifications include experience with medical imaging, radiology, or healthcare; familiarity with specific imaging modalities (chest X-ray, CT, MRI, mammography); experience with DICOM, image-report pairing, medical data de-identification, or radiology workflows; experience evaluating models across distribution shifts; familiarity with clinical validation, FDA, or HIPAA considerations; and experience with 3D vision, longitudinal imaging, or clinical decision support.
