Senior Research Scientist

DeepHealth

10d•Hybrid

About The Position

The Senior Research Scientist sits on DeepHealth's ML Capabilities team and advances research in machine learning for medical imaging, with a focus on foundation models in radiology AI. This role designs novel model architectures, develops experimental systems to test research hypotheses, and helps translate research into solutions that support clinical applications. The work draws on DeepHealth's large-scale imaging data and expert feedback loops and includes collaboration with research partners such as Imperial College London.

Requirements

PhD in Machine Learning, Computer Science, or a related field (required).
Strong publication record at top machine learning venues (for example, NeurIPS, ICML, ICLR) (required).
Expertise in deep learning and generative models (required).
Experience designing and evaluating machine learning systems (required).

Nice To Haves

Experience in medical imaging or healthcare AI.
Familiarity with multimodal learning.
Experience mentoring or leading research projects.

Responsibilities

Define and drive the scientific roadmap for foundation models in medical imaging within the ML Capabilities programme.
Track and evaluate developments in machine learning and generative AI relevant to radiology and multimodal imaging.
Design and prototype novel model architectures for multimodal medical imaging data.
Develop experimental systems to validate research hypotheses and measure progress against scientific goals.
Establish and maintain benchmarking frameworks across downstream clinical and technical tasks.
Collaborate with engineering teams to translate research advances into deployable solutions.
Lead scientific publications and conference contributions that communicate DeepHealth research results.
Engage with external research collaborators, including partners at Imperial College London.
Mentor researchers and contribute to research quality standards within the ML Capabilities team.
Document methods, experiments, and findings to support reproducibility, review, and knowledge transfer.