Home | Harish Lab — Engineering AI

We develop AI and machine-learning methods — from scientific ML and graph neural operators to foundation and generative models — and apply them across aerospace, biomedical, and civil engineering. Our work bridges modern AI research with high-impact, real-world applications.

Research Themes

Scientific Machine Learning

Graph neural operators, physics-informed ML, and surrogates for engineering systems.

Foundational AI

Foundation models, generative design, and agentic AI for engineering workflows.

Application Domains

Aerospace

Deployable space structures, tensegrities, and thin-walled composite materials.

Biomedical

Ophthalmology, cardiovascular, and soft-tissue mechanics from imaging and patient data.

Civil

Hydrodynamics — uncertainty quantification and surrogates for wave–structure interaction.

Mar 2026 New preprint: Graph neural operators for ill-posed problems in wave-structure interaction — Luo, Fourtakas & Harish
Feb 2026 New preprint: Virtual population to re-assess AAA risk using neck geometry — Nandurdikar et al.
Jan 2026 New preprint: Axial length matters: Scaling effects in retinal fundus image analysis — Li et al.
Nov 2025 Paper published in Computational Particle Mechanics: Framework for uncertainty quantification of wave–structure interaction — Luo et al.
Sep 2025 New preprint on carotid haemodynamics: In silico approach to analyse the influence of carotid haemodynamics — Sengupta et al.
May 2025 We are recruiting PhD students and postdocs — see Join Us for details.
Aug 2024 Paper published in Finite Elements in Analysis and Design: Concurrent multiscale modeling of woven fabrics — Junior et al.
Sep 2023 Paper published in Finite Elements in Analysis and Design: Alleviation techniques for volumetric locking — Harish & Matikainen

See all news →

University of Manchester

EPSRC

Optegra

INEOS Britannia

European Space Agency

Innovate UK

NHS Manchester Foundation Trust

Port of Dover