Postdoctoral Researcher in Adaptive Generative Modeling Scientific

Position: Postdoctoral Researcher in Adaptive Generative Modeling for Scientific Applications
** What You Will Do*
* The Information Sciences Group (CAI-3) in the Computing and Artificial Intelligence Division at Los Alamos National Laboratory is recruiting a postdoctoral researcher to develop and apply modern machine learning methods to complex spatiotemporal problems in physical sciences. The research will include generative modeling, with particular interest in diffusion and related models for high-dimensional, time-varying systems.

The successful candidate will join a multidisciplinary research team spanning applied mathematics, machine learning, statistics, computational physics, materials science, and accelerator physics. The research will focus on learning high-dimensional, time-varying physical systems from sparse experimental and simulation data. Application areas include three-dimensional imaging and modeling of evolving materials, virtual diagnostics for charged-particle beams, and digital twins for scientific experiments and facilities.

We are particularly interested in candidates with strong foundations in applied mathematics, physics, or statistics, who have experience in machine learning and are motivated to work closely with domain scientists. Research involve deep learning, generative models, representation learning, inverse problems, uncertainty quantification, and methods for incorporating scientific constraints or real-time observations into learned models. Prior experience with diffusion models or the specific application areas is welcome but not required.

The postdoctoral researcher will:

+ develop, implement, and evaluate machine-learning and generative models for scientific data;

+ work with large experimental and simulation datasets from multiple physical-science domains;

+ develop high-quality research software for large-scale computing platforms;

+ publish original research in leading scientific and machine learning journals and present results at conferences;

+ contribute to a general machine learning framework for reliable digital twins of time-varying physical systems;

+ collaborate with machine learning researchers, computational scientists, and experimental teams.

** What You Need*
* ** Minimum

Job Requirements:

*
* + PhD in theoretical or computational physics, applied mathematics, materials science, engineering, or a related quantitative physical-science discipline, completed within the last 5 years or to be completed soon. Candidates with a PhD in applied mathematics, computer science, statistics, or a related field must demonstrate substantial research experience in the physical sciences.

+ Strong understanding of machine learning methods as evidenced through peer-reviewed publications, presentations, scientific software, or substantial research projects.

+ Excellent programming skills and demonstrated experience using high-level languages such as Python or Julia..

+ Practical experience (outside of coursework or certifications) with modern ML tools and libraries, such as PyTorch, Tensor Flow, Keras, or JAX.

+ Experience applying machine learning to scientific, engineering, or complex quantitative data.

+ Ability to work independently and collaboratively in a multidisciplinary scientific environment.

+ Strong interpersonal, written, and oral communication skills.

** Desired

Qualifications:

*
* + Experience, substantiated by peer-reviewed publications, in generative modeling including normalizing flows, diffusion models, flow matching, stochastic interpolants, or Schrödinger bridges.

+

Experience with inverse problems, representation learning, uncertainty quantification, dynamical systems, or scientific machine learning.

+

Experience with high-performance computing, GPU programming, distributed training, parallel programming, and related paradigms.

+ Familiarity with software engineering, including testing, code version control, and continuous integration workflows.

** Work Environment:*
* *
* Work Location:

** The work location for this position is onsite and located in Los Alamos, NM. All work locations are at the discretion of management.

** Note to Applicants:*
* + The appointment is for a duration of two years, with the possibility of a third-year extension based on performance evaluation.

+ Candidates may be considered for a fellowship supported by the Center for Nonlinear Studies (CNLS). Outstanding candidates may be considered for a Director's Postdoc Fellowship or for the prestigious Richard P. Feynman, Darleane Christian Hoffman, J. Robert Oppenheimer, or Frederick Reines Distinguished Postdoc…