Human-in-the-Loop Decision-Making for Lipid Nanoparticle Reactor Design
- AI, machine learning, optimisation and reactor design
- Part of the Manchester-Cambridge UKRI AI CDT
- Fully funded 4-year PhD (£30,000 tax-free stipend)
- Home applicants
- Deadline: 4 December 2026
About the project
Lipid nanoparticles underpin mRNA vaccines and a growing class of nucleic-acid therapeutics. This project will
develop a human-in-the-loop, multi-fidelity framework for designing confined impinging-jet mixers used in lipid
nanoparticle manufacture.
Research focus
The work will combine multi-fidelity Bayesian optimisation, computational fluid dynamics, Gaussian Processes,
preference-based expert feedback, interpretable uncertainty communication, contextual-bandit decision policies
and generative reactor design.
Who should apply
Suitable for candidates with strong foundations in probabilistic methods, Gaussian Processes or Bayesian
optimisation, plus Python and mathematical maturity. Knowledge of partial differential equations, physics-based
modelling or CFD would be useful.
Manchester-Cambridge UKRI AI CDT
The project is part of the UKRI AI Centre for Doctoral Training in Decision Making for Complex Systems, which
brings together The University of Manchester and the University of Cambridge Department of Computer Science
and Technology. CDT information: https://www.ai-decisions-cdt.ac.uk/
