Geometric deep learning surrogate for hydromechanics
A machine learning model that predicts hull hydromechanics in seconds instead of hours, so designers can explore many more options.
George Drakoulas, PhD
Where physics meets AI in heavy industry
I build the bridge between physics-based simulation and production machine learning — models that make engineering decisions faster, for shipbuilding, aerospace, automotive and energy.
10× faster hydromechanics simulation
100× surrogate-model speed-up over full simulation
7 years shipping machine learning to heavy industry
Featured projects
All projectsRecent writing
All postsWhy surrogate models beat full simulation in the design loop
A four-hour simulation run is fine for validation and useless for exploration. What changes when you need an answer in seconds, and where it stops being trustworthy.
Working on something at the edge of simulation and ML?
I take on selected freelance and advisory work alongside my role in maritime R&D. If the problem is hard and physical, it's the kind I like.
Get in touch