Nathan Daniel Keilbart

Nathan Keilbart is a computational materials scientist and engineer whose work focuses on high throughput simulations, density functional theory, and data generation for machine learning and AI models.

He received his B.S. in Mechanical Engineering from Brigham Young University–Idaho in 2013, where he developed a strong foundation in applied mechanics, thermodynamics, and engineering analysis. He then pursued graduate studies at The Pennsylvania State University, earning his Ph.D. in Materials Science and Engineering in 2019. During his doctoral research, Dr. Keilbart developed and applied computational methodologies based on density functional theory (DFT) to predict the energy storage performance of pseudocapacitors. His work centered on understanding how atomic scale structure and chemistry influence charge storage mechanisms, with the goal of guiding the design of improved materials for electrochemical energy storage.

After completing his Ph.D., Dr. Keilbart joined Lawrence Livermore National Laboratory (LLNL) as a postdoctoral researcher in the Physical and Life Sciences Directorate (PLS), Materials Science Division (MSD), within the Quantum Simulations Group (QSG). In this role, he contributed to a diverse portfolio of simulation driven materials research. His postdoctoral projects included:

• Corrosion prediction and prevention
  Using atomistic and electronic structure methods to understand corrosion mechanisms and support the development of more robust materials and coatings.
• High throughput workflow development
  Designing, implementing, and deploying automated computational workflows to perform large scale DFT and related simulations, enabling systematic exploration of materials spaces.
• Machine learning for materials and spectroscopy
  Assisting in the development and training of machine learning models to predict materials properties and spectral signatures, integrating physics based simulations with data driven approaches.

Following his successful postdoctoral work, Dr. Keilbart converted to a staff scientist position at LLNL. As staff, he has focused on leveraging high throughput computational workflows to generate high quality, large scale datasets that accelerate the development and validation of AI and machine learning models. His efforts emphasize reproducible, automated, and scalable simulation pipelines that can support a wide range of scientific and national security applications.

His recent and ongoing work includes applying these workflows to:

• Epidemiology simulations
  Supporting disease spread modeling and scenario exploration through large ensembles of computational simulations, aimed at improving predictive capabilities and informing public health decision making.
• Energetic materials
  Studying the behavior, stability, and properties of energetic materials using high throughput calculations, with an emphasis on generating data that can be used to train SI models for safety, performance prediction, and materials discovery.
• Power grid modeling
  Contributing to simulations and data generation efforts that support modeling, analysis, and resilience studies of power grids, with a focus on scalable workflows that can explore many scenarios and parameter spaces.

Across these areas, Dr. Keilbart’s scientific interests lie at the intersection of computational materials science, large scale simulation, and SI driven modeling. He specializes in building and applying high throughput computational frameworks that bridge fundamental physics based methods, such as DFT, with modern data science tools to address complex, multidisciplinary problems.