Artificial Intelligence for Materials Postdoctoral Research Associate

About the role

The Materials Physics and Application Division - Center for Integrated Nanotechnologies (MPA-CINT) and Theoretical Division - Physics and Chemistry of Materials (T-1) at Los Alamos National Laboratory seeks a highly motivated post-doctoral candidate, in the areas of machine learning and AI for materials science, with an emphasis on learning from materials datasets to establish structure-property relationship.

Responsibilities

• Work in Group CINT of the Materials Physics and Application Division and collaborate with a larger group of scientists and postdocs from across organizations.
• Participate in a larger project focused on ML/AI models for regression, classification, and uncertainty-aware prediction to connect microstructure/defect descriptors (from simulations and/or imaging-derived features) to materials properties and performance.
• Integrate with the project team broadly and coordinate with experimentalists.
• Contribute to the development and use of atomic-scale descriptors for learning from atomistic datasets (SFD, ACE, MTP, SNAP), including feature construction, and integration into ML workflows.
• Develop and use ML/AI models at scale on GPU-accelerated HPC systems, including managing large datasets/workloads and performance-aware workflows.
• Conduct original scientific research through peer-reviewed publication records.
• Communicate results through publications in peer-reviewed journals and presentations at prominent conferences.

Requirements

• Demonstrated expertise in one or more of the following: Machine learning for materials data, including regression and classification on materials and microstructure datasets.
• Materials modeling experience relevant to microstructure and mechanics (e.g., MD/DFT, phase-field, crystal plasticity, microstructure-property relationships).
• Uncertainty quantification (UQ) and model reliability, including approaches such as calibrated probabilistic prediction, Bayesian/ensemble methods.
• Strong programming skills in Python (e.g., NumPy/SciPy, pandas,) and ML frameworks such as scikit-learn, PyTorch, JAX, and TensorFlow.
• Excellent communication skills (both oral and written).

Qualifications

• A STEM PhD in areas such as Materials Science, Computational Physics, Engineering, or related fields, completed within the last five years or soon to be completed.
• Solid Background in materials science and engineering.
• Experience training ML/AI models at scale on GPU-accelerated HPC systems, including managing large datasets/workloads and performance-aware workflows.
• Ability to adapt to new requirements for projects and be flexible enough to learn new areas of research as needed.
• Ability to work effectively as a part of a team in a multi-disciplinary environment and interact with people with a variety of expertise.