Research Fellow, School of Computing

Interested applicants are invited to apply directly at the NUS Career Portal. Please note your application will only be processed if you apply via NUS Career Portal.

NUS Career Portal link: https://careers.nus.edu.sg/job/Research-Fellow%2C-School-of-Computing/34714-en_GB/st=7B0866D297C16F918CB3FD5E2986A487855A8AB2

We regret that only shortlisted candidates will be notified.

Job Description

The Research Fellow will serve as the Lead Researcher for Work Package 1 (WP1), An Equivariant Foundation Model for Crystal Structures, under the project Advancing Generative Models for Materials: Unified AI and Mathematical Foundations.
The candidate will lead the development of novel, scalable, and symmetry-aware generative AI models for crystalline materials. His primary responsibilities will include designing and implementing equivariant Transformer architectures, developing anchor-based flow-matching techniques, and incorporating rotational, translational, and periodic symmetries into the generative framework.

He will oversee large-scale model training and evaluation using materials datasets, including MP20, MP-Large, and OMC25, and optimize computational performance using dedicated GPU infrastructure. He will also evaluate the validity, diversity, novelty, and physical stability of generated crystal structures and coordinate their validation with collaborators.

Additionally, the candidate will collaborate closely with researchers across the project's other Work Packages, particularly WP2, to integrate the developed foundation model into conditional materials generation. He will contribute to scientific publications, international conference presentations, and the release of reproducible, open-source software and pre-trained models.

Qualifications

• PhD in Computer Science, Artificial Intelligence, Applied Mathematics, Physics, or a closely related discipline.
• Strong research background in deep learning, particularly geometric deep learning, graph neural networks, Transformers, and generative AI models.
• Solid mathematical foundations in linear algebra, geometry, probability, and optimization, with an understanding of symmetry-aware and equivariant machine learning.
• Excellent programming skills in Python and experience with deep learning frameworks such as PyTorch, including GPU-based model development and training.
• Demonstrated ability to conduct independent, high-quality research, develop novel machine learning methodologies, and publish findings in reputable international journals or conferences.
• Experience with large-scale machine learning, foundation models, flow matching, or scientific computing would be highly advantageous.
• Knowledge of computational physics, materials science, or AI applications in scientific discovery would be an advantage.
• Strong analytical and problem-solving skills, with the ability to translate mathematical concepts into efficient computational implementations.
• Excellent written and verbal communication skills in English, with the ability to work independently and collaboratively within an interdisciplinary research team.
• Strong motivation to pursue innovative research at the intersection of artificial intelligence, mathematics, and physical sciences.