Physics-Informed ML Engineer for Ship Design & Hydrodynamics

The Methodist Church in Britain; London Borough of Southwark; Hybrid; GBP 35,000 - 55,000; Full time

Benefits offered by this job

£6,000 personal development budget

Hybrid working may be available

Job summary

Compute Maritime Ltd in collaboration with the University of Southampton is seeking a Machine Learning Engineer for a Knowledge Transfer Partnership to embed physics-informed AI into NeuralShipper, a vessel design platform. You will build fast ML models for ship resistance and propulsion, and extend to wind-assisted propulsion and rigid sails.

You will work with industry stakeholders and use CFD/benchmark data to validate tools, developing scalable software solutions for commercial deployment.

Qualifications

• Experience applying ML/DL to engineering or physical systems.
• Experience with engineering simulation tools such as STAR-CCM+ for CFD, hydrodynamics or vessel performance.

Responsibilities

• Manage work packages to turn research into commercially viable solutions.
• Develop and validate fast physics-informed models for ship resistance, propulsion and energy efficiency using CFD and benchmarks.
• Design and implement multidisciplinary optimisation methods and integrate them into NeuralShipper as scalable software tools.
• Extend NeuralShipper to wind-assisted propulsion and rigid sail systems.
• Collaborate with industry stakeholders to validate tools against practical design requirements.

Skills

Python

Machine learning

Deep learning

CFD

Education

MSc/MEng/PhD in related discipline

Tools

STAR-CCM+

PyTorch

TensorFlow

JAX

Job description

Compute Maritime Ltd in collaboration with the University of Southampton is seeking a Machine Learning Engineer for a Knowledge Transfer Partnership to embed physics-informed AI into NeuralShipper, a vessel design platform. You will build fast ML models for ship resistance and propulsion, and extend to wind-assisted propulsion and rigid sails.

You will work with industry stakeholders and use CFD/benchmark data to validate tools, developing scalable software solutions for commercial deployment.