Location: London (hybrid working may be available)
A Computational Ship Hydrodynamics and Design Optimisation specialist is required to work on an ambitious and novel project to embed physics informed generative AI tools within a marine vessel concept, generation and evaluation platform.
This will be part of a Knowledge Transfer Partnership (KTP), which is a collaborative project between Compute Maritime Ltd and the University of Southampton.
Find out more about Knowledge Transfer Partnerships here.
Compute Maritime Ltd is a London-based deep-tech company bringing intelligence to the core of the global shipbuilding industry through generative artificial intelligence (AI) and high-performance computing.
Through its proprietary technologies, most notably NeuralShipper, the company is building the first AI-native maritime design ecosystem, offering end-to-end solutions across the vessel lifecycle, from early concept design to operational optimisation.
The Machine Learning Engineer will be required to undertake the following:
- Translate and embed research into commercially viable solution by managing a series of work packages.
- Develop and validate fast, physics-informed models for predicting ship resistance, propulsion performance and energy efficiency using CFD and benchmark data.
- Design and implement multidisciplinary optimisation methods, integrating them into NeuralShipper as robust and scalable software tools for automated vessel design improvement.
- Extend NeuralShipper’s capabilities to wind-assisted propulsion and rigid sail systems, working with industry stakeholders to validate the tools against practical design requirements.
The successful Machine Learning Engineer will have the following skills, experience and attributes:
- MSc/MEng or PhD (desirable) in Machine Learning, AI, Computational Fluid Dynamics, Hydrodynamics, Optimisation, or a related discipline.
- Experience of applying machine learning and deep learning to engineering or physical systems.
- Strong scientific programming skills in Python, with experience in C++, MATLAB, or similar languages desirable.
- Experience with a deep learning framework such as PyTorch, TensorFlow, or JAX (desirable).
- Experience with engineering simulation tools relevant to CFD, hydrodynamics, or vessel performance, such as STAR-CCM+.
- Understanding of naval architecture, ship hydrodynamics, vessel performance, or design analysis.
- Experience in physics-informed machine learning, surrogate modelling, generative AI, or design optimisation would be desirable.
- An entrepreneurial mindset and a willingness to build commercial acumen alongside technical strengths.
Personal development: A separate £6,000 budget is available over the duration of the KTP for relevant training, conferences and professional memberships.