Postdoc: Scientific ML for Excited‑State Chemistry

LBL

California (MO)

Hybrid

USD 99,000 - 114,000

Full time

9 days ago

Application generator

Turn this role into an interview — a resume and cover letter built around what this employer wants.

Get past ATS filters

Job summary

Lawrence Berkeley National Laboratory (Berkeley Lab) seeks a Postdoctoral Researcher in Scientific Machine Learning & Computational Chemistry to advance ML methods for excited-state molecular dynamics and computational chemistry. You will develop physics-informed, symmetry-aware models, evaluate them on energy-relevant systems, and connect them to nonadiabatic workflows and open-source tools.

The role emphasizes hands-on software development, rigorous validation, and collaboration across diverse

Qualifications

PhD in Computer Science, Computational Science, Chemistry, Physics, Applied Mathematics, Materials Science, Chemical Engineering, or related field.
Strong Python with PyTorch or JAX.
Experience applying ML to scientific/atomistic data.
Knowledge of graph neural networks and symmetry-aware models.
Ability to develop reliable research software in Linux with version control.

Responsibilities

Conduct fundamental research in physics-informed and symmetry-aware ML for nonadiabatic excited-state dynamics.
Develop equivariant GNNs and related architectures for multi-state PES learning.
Design data-generation, active learning, transfer learning, and uncertainty analyses to reduce data costs.
Implement and maintain open-source Python/JAX software; integrate models with nonadiabatic workflows and simulation interfaces.
Publish results in peer-reviewed journals and contribute to software releases and collaborations.