Postdoctoral Researcher

**Job Description**
The successful candidate will conduct research on cutting-edge machine learning methods, specifically focusing on generative and predictive approaches for materials discovery. This fully funded postdoctoral position at DTU Compute involves advancing neural network-based methods within the “AI-driven materials optimization for light trapping in thin-film solar cells” project, in close collaboration with DTU Physics and DTU Nanolab. The project aims to develop neural diffusion techniques for designing materials with targeted optical properties and create multi-fidelity predictive models integrating quantum simulations and experimental data, ultimately enabling iterative improvement through a closed-loop framework.

**Skills & Abilities**
• Proven experience with graph neural networks, transformers, or similar neural network methods applied to materials or molecules.
• Proven experience with implementing machine learning methods in Python and Pytorch.
• Experience with deployment on GPU clusters.
• Strong publication record in graph neural networks, geometric deep learning, representation learning, machine learning-based molecular/materials discovery, diffusion-based generative models, or other related fields.
• High level of motivation and creative problem-solving skills.
• Excellent communication and writing skills in English.

**Experience**
Experience Required:
• Proven experience with graph neural networks, transformers, or similar neural network methods applied to materials or molecules.
• Proven experience with implementing machine learning methods in Python and Pytorch.
• Experience with deployment on GPU clusters.
Other:
• Strong publication record in graph neural networks, geometric deep learning, representation learning, machine learning-based molecular/materials discovery, diffusion-based generative models, or other related fields.

Note: We’ve analyzed the actual job post using AI, for more details visit the original job post by clicking on “Apply Now”!