Senior AI Engineer on Foundational Models for Healthcare

Location: St Thomas' Campus

Contact details: Jorge Cardoso. m.jorge.cardoso@kcl.ac.uk

About us

The School of Biomedical Engineering & Imaging Sciences (BMEIS), part of King’s College London, is driven by a commitment to improve the way healthcare is delivered through the use of advanced engineering.  

BMEIS is diverse and talented group working across the whole MedTech sector, advancing research, innovation and teaching progress through a shared mission of engineering better health for patients worldwide. The School’s state-of-the-art labs and clinical-research facilities are embedded in St Thomas’ Hospital to ensure our research projects are fully aligned with current clinical practice. Long-term collaborations with global MedTech companies and new partnerships with innovative start-ups ensure multiple pathways to translation.

About the role

The AI Centre, led by King’s College London and Guy's and Thomas’ NHS Foundation Trust, is a consortium of multiple NHS Trusts, Universities and UK and multi-national industry partners that has developed platforms allowing the development and deployment of sophisticated artificial intelligence algorithms from NHS data, to provide tools for clinicians to speed up and improve diagnosis and care across several patient pathways.

As part of these platforms, the AI centre created the Federated Learning Interoperability Platform (FLIP), which helps link data from multiple NHS Trusts and enable AI at scale. This platform is now being extended beyond tertiary care to the London Sub-national Secure Data Environment (SNSDE), enabling extended use of patient data across the care continuum.

A very significant amount of data is available as part of these platforms, motivating the development of vision-language foundational models to not only learn from these data, but also to serve as a foundation and support future task-specific AI research developments. This role will focus exactly on building such a model foundation, ie. training a large vision (medical imaging) and language (radiological reports) model that allows us not only to interact with the vision data via a language interface (visual QA), tackle several downstream tasks (generating reports from images, images from reports, and calculating discrepancies between images and reports), but also demonstrating that the latent representations created by such models are rich enough for downstream tasks and can be used as a foundation to develop smaller and more task-specific models. As part of this algorithmic development journey, the applicant is also expected to make fundamental algorithmic contributions in building foundational models, contribute towards open source software (MONAI Core and Generative), and improve our ability to use healthcare data at scale.

This is a full time post, and you will be offered a fixed term contract until 31 March 2026.

About You

To be successful in this role, we are looking for candidates to have the following skills and experience:

Essential criteria

• PhD awarded in Mathematics, Engineering or Computer Science or near completion.
• Higher language computer programming (python)
• General machine learning experience (vision, structured data, NLP)
• Machine learning libraries (SciKit Learn, pytorch,…)
• Scientific / Medical Writing
• Ability to work calmly under pressure and act on initiative.
• Interest in medical imaging

Desirable criteria

• Knowledge of software development cycles
• Data analysis packages (Pandas, SciPy/Statsmodels, etc.)
• Visualisation tools (Paraview, OHIF, etc)
• HPC computing services (SGE-like batch-queuing system, Kubernetes, Docker, Containers)
• Independent researcher

Advert information