Postdoctoral Research Associate - AI​/ML Accelerated Theory, Modeling and Simulation Materials

Pay Competitive

Employment type:

Full-Time

• Req#: 9474
• Requisition

Overview:

Oak Ridge National Laboratory is the largest US Department of Energy science and energy laboratory, conducting basic and applied research to deliver transformative solutions to compelling problems in energy and security.

We are seeking a Postdoctoral Research Associate who will support the Nanomaterials Theory Institute, Theory and Computation Section in the Center for Nanophase Materials Sciences Division (CNMS), Physical Sciences Directorate (PSD) at Oak Ridge National Laboratory (ORNL). The Postdoctoral Research Associate will support research directed toward developing a computational approach to perform AI/ML accelerated predictive theoretical simulations that can guide nanoscale experiments to fabricate nanostructures with desired functionalities, in a directed manner, thereby enabling autonomous experimental synthesis/manipulation of functional nanomaterials.

Focus will largely be in developing ML-algorithms for computational discovery of new materials with targeted functionalities that can be fabricated at the atomic-scale using automated microscopy (STM, STEM and SPM) and synthesis (PLD and MBE) techniques, based on training data obtained from predictive modeling. Predictive theoretical modeling will be achieved by combining scalable density functional theory (DFT) approaches (such as real-space DFT, DFTB), beyond-DFT approaches for solids (such as GW, DMFT, QMC), and reactive force-field methods, using advanced automated workflows and AI/ML algorithms, and incorporate experimental feedback to improve theory.

As a Postdoctoral Research Associate, you will contribute to research in these areas by bridging state-of-the-art atomistic simulation methods, as indicated above, and nanoscale experiments with domain-informed AI/ML algorithms. In addition to fundamental science discovery, the research will pursue development of automated workflows and ML-approaches that allows integration of different theory & simulation protocols along with autonomous experiments. The research is designed to provide opportunities for development of your experience and scientific vision.

You will also work closely with scientists at CNMS who are experts in developing/applying different theoretical approaches as well as those developing novel ML-approaches for automation of nanoscale experimental probes, as mentioned above.

Major Duties/Responsibilities:

• Conduct research to develop and apply AI/ML algorithms that can result in accelerated materials discovery and understanding using predictive theoretical modeling
• Develop and apply automated workflows to perform reactive atomistic simulations employing new ML-based statistical sampling approaches that allow larger length-scales and long time-scale simulations
• Create and maintain datasets in computational databases on in-house data storage resources working closely with ORNL’s workflow and data management scientists
• Develop AI/ML approaches to incorporate experimental feedback and improve theoretical predictions
• Meaningfully collaborate with experimental groups at CNMS and actively contribute to core science projects
• Report and publish scientific results in peer-reviewed journals in a timely manner
• Present results at international scientific conferences and meetings

Basic Qualifications:

• A PhD in Physics, Materials Science, Chemistry, or a related field completed within the last 5 years
• Strong background in developing and/or applying electronic structure methods
• Sound understanding of ML concepts and hands-on experience with open-source AI/ML packages (such as pytorch, scikit-learn, tensor flow, JAX etc.)
• Strong publication list using electronic structure methods

Preferred Qualifications:

• Good grasp of concepts in solid-state physics, topological condensed matter theory, and/or quantum chemistry
• Familiarity with correlated electronic structure methods, particularly QMC or GW or DMFT, and/or reactive force-field methods
• Strong demonstrated background in coding for data analysis using Python, Julia…