Research Fellow (Plasma Physics for Fusion Energy)

The School of Physical and Mathematical Sciences (SPMS) at NTU Singapore hosts research and education activities in two divisions: Division of Mathematical Sciences (MAS) and Division of Physics and Applied Physics (PAP). MAS covers diverse topics ranging from pure mathematics to the applications of mathematics in cryptography, computing, business, and finance. PAP covers many areas of fundamental and applied physics, including quantum information, condensed matter physics, biophysics, and photonics. Over the years, SPMS has attracted talented individuals from around the world and Singapore to join as scientific leaders and researchers.

Job responsibilities

The project aims to advance the use of machine learning techniques to model and understand plasma turbulence in magnetically confined fusion plasmas. The Research Fellow will be employed and based at NTU, and is expected to travel overseas to collaborate with scientific teams at CEA, France.

The holder will:

• Apply advanced machine learning algorithms, including deep neural networks and physics-informed neural networks, to analyse large datasets from gyrokinetic and fluid simulations of plasma turbulence

• Develop and train reduced-order models that capture the essential dynamics of turbulent transport in tokamak plasmas, enabling fast predictive capability

• Benchmark AI-driven models against experimental observations and conventional simulations to validate performance across different plasma scenarios

• Collaborate with international partners to integrate AI models into broader fusion research programmes, sharing data, methodologies, and validation strategies

Job requirements

• PhD in Physics, Applied Mathematics, Computational Science, or a related field

• Strong background in machine learning, particularly in the development and application of neural networks for scientific data analysis

• Familiarity with plasma physics, especially turbulence and transport, is desirable

• Proficiency in scientific programming and data analysis, including the use of modern tools and libraries (e.g. Python, TensorFlow/PyTorch, Scikit-learn)

• Demonstrated experience in applying machine learning or data-driven methods to physical systems, preferably in the context of fusion research

• Proven track record of scientific productivity through publications and/or software contributions

• Strong communication skills for presenting results and writing scientific publications

The College of Science seeks a diverse and inclusive workforce and is committed to equality of opportunity. We welcome applications from all and recruit on the basis of merit, regardless of age, race, gender, religion, marital status and family responsibilities, or disability.

We regret that only shortlisted candidates will be notified.