PostDoctoral Researcher in Pediatric AI Clinician Development

Our vision is to develop our semi-autonomous AI system for saving lives in intensive care, the “AI clinician”. Our team won three prestigious research grants including the UKRI Turing AI Fellowship to Prof Aldo Faisal,  Rosetrees Interdisciplinary Award, British Medical Association Foundation John Molton Award towards Dr Padmanabhant Ramnarayan, M.D. our program is funding a number of positions to grow our AI Clinician team and apply to paediatric intensive care. The AI Clinician is informed by myriads of patient records that is using Generative AI from operational data to learn optimal pharmaceutical treatment policies for critically ill patients that we pioneered (Komorowski et Faisal, 2018, Nature Medicine; Gottesmann et al, 2019, Nature Medicine). On the clinical side of the AI Clinician we are a team spanning world-leading intensive care clinicians Padmanabhan Ramnarayan (paediatric ICU) and Anthony Gordon (adult ICU) and Aldo Faisal (Director, UKRI Centre in AI For Healthcare) at Imperial College London. Our pioneering adult ICU system is now in clinical trials and being deployed in 4 intensive care units across London as a healthcare flagship in Digital Therapeutics.

Our 4 new posts are going to be embedded in our AI Clinician team spanning a vibrant group of proactive research ranging from machine learning (reinforcement learning and generative AI), explainable AI and intensive care clinicians to experts supporting the human factors, patient & stakeholder engagement that will enable us to deploy this  Digital Therapeutics  flagship program in NHS hospitals in the course of these posts.

The clinical aims of the project are to expand the expertise domain and capabilities of the AI Clinician to broaden the patient base by including paediatric intensive care units (PICUs) to help children and  to broaden its action space to encompass ventilation decisions and control as well as fluid management.

The 4 roles span the domain from foundations of machine learning to data engineering for the deployment of the system in our trial hospitals. The 4 posts are as follows 1. PostDoc advancing off-line Reinforcement Learning and Decision Transformer research that underpins the AI Clinician , 2. PostDoc  Developing the paediatric AI Clinician model, 3. PostDoc Developing the Data Science of the closed-loop operation of the paediatric AI Clinician., 4. (RA) Research software engineering of the paediatric AI Clinician decision support systems including integration of the live data feeds of the electronic patient records. Clinical ethics  and  IT approvals are already in place, and follow our successful adult AI Clinician work.

Posts 2.-4. are jointly supervised by Aldo Faisal and Padmanabhan Ramnarayan. The overall research environment is corresponding to our ambition, the Departments where our research group is based has been consistently ranked as the top Departments in the country, and Imperial College has been consistently ranked in the top-10 universities world-wide. Imperial College boasts a world-leading ecosystem of AI for Healthcare research across the breadth of both AI and Medicine, and hosts among other things the pioneering UKRI Centre in AI for Healthcare as well as the AI network, comprising over 200 faculty and 1000 Research Associates and PhD Students at Imperial that work in AI.

The PostDoctoral posts 1. Will advance machine learning methods for the AI Clinician with their immediate application to clinical intervention this comprises reinforcement learning (e.g. Li & Faisal AAAI, 2021; but also including decision transformers) and/or causal learning methods (including dynamic treatment regimes). It offers the unique opportunity to develop core machine learning theory and see it evaluated through our team. The PostDoctoral posts 2. and 3. will be working on creating the paediatric AI Clinician, post 2. Will focus on building and training the Paediatric AI Clinician model, and post 3. Will be focussed on incorporating the clinical data sets we already hold and the clinical evaluation. The research software engineer 4. will be developing the NHS IT interfacing that pulls live data and historical data to process it in our secure data enclave.

• PDRA posts 1.-3. only: a PhD (or near to completion) or equivalent in a relevant field*


• Relevant computer science/engineering/mathematics/physics or exceptional biomedical undergraduate background


• Experience in the field of (one more multiple) Digital Healthcare, Reinforcement Learning, Machine Learning, Robotics, Computational Neuroscience, Recommender Systems Neural/Behaviour Data Analysis, Time Series Analysis, or relevant related backgrounds in control systems or Mathematics


• Strong analytical mathematical skills


• Strong programming skills


• Strong machine learning skills.


• Willingness to work in a highly dynamic & interdisciplinary team of technical, medical and human-centered engineering experts

The Imperial jobs site will be temporarily unavailable from 16.00pm Friday 23rd February to 12.00pm Wednesday 28th February as we move to a new Job site platform. 
We will not be accepting applications or advertising vacancies during this period.
Applications started, not yet submitted by 16:00 23rd February will need to be started again on the new platform.
For further information and how to submit an application or set up job alerts again please visit our FAQs for new job site platform transition

Please note the URL (https://www.imperial.ac.uk/jobs/ ) will be the same.

This is a full time, fixed term post, with a possibility of extension.

Research Associate: PhD (or equivalent) in Reinforcement Learning, Machine Learning, Robotics, Recommender Systems, Computational Cognitive Science, Computational Neuroscience,  Neural/Behaviour Data Analysis, Time Series Analysis, or relevant related backgrounds in control systems or Mathematics.

Research Assistant: Near completion of a PhD (or equivalent) in Reinforcement Learning, Machine Learning, Robotics, Recommender Systems, Computational Cognitive Science, Computational Neuroscience,  Neural/Behaviour Data Analysis, Time Series Analysis, or relevant related backgrounds in control systems or Mathematics

*Candidates who have not yet been officially awarded their PhD will be appointed as a Research Assistant within the salary range £40, 694 - £43, 888 per annum.*

Our preferred method of application is online via our website. Please click ‘apply’ below or go to http://www3.imperial.ac.uk/employment . (Select “Job Search” then enter the job title or vacancy reference number into “Keywords”).

For any informal enquiries regarding the post please do not hesitate to contact Aldo ([email protected] , Subject: [Post-Doc AI Clinician])

Alternatively, if you are unable to apply online, please email Mica Jones on [email protected] to request an application form.

Imperial Managers lead by example.

Committed to equality and valuing diversity. We are also an Athena SWAN Silver Award winner, a Stonewall Diversity Champion, a Two Ticks Employer, and are working in partnership with GIRES to promote respect for trans people