Postdoctoral Fellow - Generative AI Researcher - 3D Defect Synthesis King Abdullah University of Science and Technology (KAUST) in Saudi Arabia

King Abdullah University of Science and Technology (KAUST) is recruiting a Postdoctoral Fellow - Generative AI Researcher - 3D Defect Synthesis in the Physical Science and Engineering Division, within Material Science and Engineering. This postdoctoral role focuses on a high-impact research problem at the intersection of generative AI, volumetric imaging, and industrial non-destructive testing.

The project addresses a major bottleneck in AI-powered inspection: real defects in batteries, consumer electronics, and structural components are rare, expensive to create, and difficult to reproduce at scale. The successful candidate will develop generative models that can synthesize geometrically and physically plausible defects directly into 3D CT and X-ray volumetric data, producing synthetic datasets for training high-fidelity detection systems without relying on large numbers of real defective samples.

Research topics include diffusion, GAN, or NeRF-based defect synthesis engines; physics-aware rendering that respects X-ray attenuation, Hounsfield unit gradients, CT reconstruction artifacts, and material contrast; multi-scale CT dataset pipelines spanning micro-CT, CBCT, and X-ray projections; and closed-loop integration with YOLO, segmentation, and anomaly detection workflows. The advertised scales range from centimeter-level structural failures down to nanometre-level material anomalies, including cracks, delamination, voids, inclusions, porosity, dendrite growth, thin-film defects, and lattice-level anomalies.

Applicants should hold a PhD in Computer Vision, Medical Imaging, Applied Machine Learning, or a closely related field. The posting asks for hands-on experience with generative models applied to 3D or volumetric data, strong knowledge of CT or X-ray imaging pipelines, and experience with anomaly detection or defect detection models. Proficiency in Python and PyTorch is required, while experience with MONAI, ASTRA, SIRT/FDK reconstruction, industrial NDT, materials science, semiconductor inspection, battery inspection, HDF5/Zarr data handling, and GPU-accelerated volumetric processing are all advantageous.

The role offers an internationally competitive, tax-free salary, on-campus housing, comprehensive health insurance, annual travel allowance, and access to world-class CT, micro-CT, and imaging facilities in a vibrant international research environment at KAUST in Thuwal, Saudi Arabia.

Applications are submitted via Interfolio only. Candidates must provide a CV, a one-page statement of purpose, and answers to five required screening questions covering debugging experience, synthetic-to-real transfer challenges, a recent paper, end-to-end system deployment, and code evidence. The post is reviewed on a rolling basis and remains open until filled.