Postdoc: Scientific ML for Molecular Dynamics & Chemistry

Lawrence Berkeley National Laboratory; Berkeley (CA); Presencial; USD 99.000 - 114.000; Jornada completa

Descripción de la vacante

Lawrence Berkeley National Laboratory (Berkeley Lab) seeks a Postdoctoral Researcher to advance scientific machine learning for computational chemistry across excited-state problems. You will develop symmetry-aware ML models and integrate them with molecular dynamics workflows, delivering robust, open-source implementations.

The role emphasizes physics-informed ML, equivariant GNNs, active learning, and collaboration with teams across DeepMD, NEXMD, i-PI, MixPI, and CP2K interfaces.

Formación

• PhD in Computer Science, Computational Science, Chemistry, Physics, Applied Mathematics, Materials Science, Chemical Engineering, or related field within the last 3 years.
• Strong Python programming skills and experience developing, training, and evaluating neural-network models with PyTorch or JAX.
• Knowledge of graph neural networks, geometric deep learning, invariant or equivariant models for molecular/physical systems.
• Ability to develop reliable research software in Linux with version control, testing, documentation, and reproducible workflows.

Responsabilidades

• Conduct fundamental research in physics-informed and symmetry-aware machine learning for nonadiabatic excited-state molecular dynamics.
• Develop and evaluate equivariant graph neural networks and related architectures learning adiabatic PES, energies, gradients or forces and nonadiabatic couplings.
• Design data-generation and training workflows including sampling, active learning, transfer learning, validation, and uncertainty analysis.
• Implement, test, document, and maintain research software; integrate models with surface-hopping and nonadiabatic workflows and simulation interfaces.
• Publish results in peer-reviewed journals and mentor students and researchers.

Conocimientos

Python

PyTorch

Graph neural networks

JAX

Linux

GPU computing

Open-source software

Collaborative research

Educación

PhD degree

Herramientas

DeepMD

NEXMD

i-PI

MixPI

NWChem

CP2K

Descripción del empleo

Lawrence Berkeley National Laboratory (Berkeley Lab) seeks a Postdoctoral Researcher to advance scientific machine learning for computational chemistry across excited-state problems. You will develop symmetry-aware ML models and integrate them with molecular dynamics workflows, delivering robust, open-source implementations.

The role emphasizes physics-informed ML, equivariant GNNs, active learning, and collaboration with teams across DeepMD, NEXMD, i-PI, MixPI, and CP2K interfaces.