Postdoc: MLIP & Active-Learning MD on HPC

Oak Ridge National Laboratory; Oak Ridge (TN); On-site; USD 65,000 - 90,000; Full time

Job summary

Oak Ridge National Laboratory seeks a Postdoctoral Research Associate to advance active learning workflows for generating datasets and training ML interatomic potentials. You will develop MLIPs for MD simulations and optimize large-scale computations on HPC systems, collaborating across disciplines to publish results.

The role emphasizes Python-based software, GPU-aware workflows, and using tools such as LAMMPS, ASE, PyTorch, with a focus on scalable atomistic modeling at ORNL.

Qualifications

• PhD completed within the last five years.
• Experience with molecular dynamics (MD) simulations.
• Experience with MLIPs or atomistic ML methods.
• Experience with Python and scientific software development.
• Knowledge of ML concepts including training, validation, and neural networks.
• Experience in Linux-based HPC environments.

Responsibilities

• Develop MLIPs for molecular dynamics simulations.
• Develop and optimize automated workflows for data generation and active learning.
• Implement and optimize large-scale MD simulations on HPC resources.
• Evaluate MLIP architectures, pretrained models, and mission-specific potentials.
• Develop and maintain scientific software in Python with modern practices.
• Collaborate with researchers across domains and publish results.

Skills

Python

MD simulations

HPC experience

Machine learning concepts

Communication skills

Education

Ph.D. in Chemistry, Physics, Materials Science, Chemical Engineering, Computer Science, Applied Mathematics or related

Tools

LAMMPS

ASE

PyTorch

GROMACS

OpenMM

TorchSim

Job description

Oak Ridge National Laboratory seeks a Postdoctoral Research Associate to advance active learning workflows for generating datasets and training ML interatomic potentials. You will develop MLIPs for MD simulations and optimize large-scale computations on HPC systems, collaborating across disciplines to publish results.

The role emphasizes Python-based software, GPU-aware workflows, and using tools such as LAMMPS, ASE, PyTorch, with a focus on scalable atomistic modeling at ORNL.