Postdoctoral Research Associate Machine Learning
Brookhaven National Laboratory
Upton, NY
DescriptionNational Synchrotron Light Source II (NSLS-II)’s Photon Science Division aims to enable world-class research by delivering exceptional scientific tools to a broad user community that requires large-scale facilities. Enhanced data analysis for x-ray scattering data, independent of the scientific sample, is essential to this mission. The Coherent Soft X-ray (CSX) and Coherent Hard X-ray (CHX) beamlines at NSLS-II are partnering with Brookhaven National Laboratory’s Computer Science Initiative (CSI) to test existing or develop new machine learning algorithms to that reduce the required expertise in data reduction and analysis. X-ray Photon Correlation Spectroscopy (XPCS), which uses a time series of x-ray scattering patterns from a given sample, will be the focus of this effort.

The chosen candidate will work closely with CSI and NSLS-II scientists to develop machine learning algorithms that can distinguish between "good" time series data sets and those containing systematic error, quantify systematic error effects to salvage data, and use machine learning to extract physically meaningful quantities from large data sets. These capabilities will be applicable to equilibrium and non-equilibrium scientific samples probed by x-rays, allow external researchers to spend more time understanding results and planning experiments, and reduce the required expertise and knowledge to perform these experiments. For reference, refer to publicly available work: Scientific Reports 11, 14756 (2021) and arXiv preprint arXiv:2201.07889 (2022).
RequirementsEssential Duties and Responsibilities: • Develop machine learning and artificial Intelligence (AI) methods for coherent scattering time series data sets (e.g., XPCS) with the end goal of reliably and accurately quantifying observed timescales of sample dynamics regardless of the data quality • Develop machine learning and AI methods that can integrate with NSLS-II’s data pipeline (beamline servers and dedicated HPC-like resources) • Collaborate with a variety of researchers, e.g., experimental scientists, mathematicians and theorists, data scientists and engineers, and computational scientists, to complete project deliverables • Present and publish results. Required Knowledge, Skills, and Abilities: • Doctorate (Ph.D.) in computational science, applied mathematics, physical science, or related field • Demonstrated record of scientific excellence (course work, publications, talks, or documented software projects) relevant to duties and responsibilities • Ability to work collaboratively with individuals from diverse scientific fields • Experience writing scientific data analysis software in Python. Preferred Knowledge, Skills, and Abilities: • Experience with object-oriented software development in C++ • Experience with common machine learning methods (e.g., convolutional neural networks) • Experience with common machine learning software libraries (e.g., PyTorch) • Experience in post-processing, model development, and analysis of large data sets • Ability to understand data from a physically meaningful point of view • Demonstrated software development, preferably in a collaborative and distributed manner.
Company DescriptionBrookhaven National Laboratory ( is a multidisciplinary laboratory that delivers discovery science and transformative technology to power and secure the nation’s future. The Lab is primarily supported by the U.S. Department of Energy’s (DOE) Office of Science. Brookhaven Science Associates, a partnership between Battelle and The Research Foundation for the State University of New York on behalf of Stony Brook University, operates and manages the Laboratory for DOE. Brookhaven Lab's Computational Science Initiative ( excels at integrating computer science, applied mathematics, data science, quantum and computational science with broad domain expertise. These diverse efforts now support CSI's stewardship of the Lab-wide Human-AI-Facility Integration (HAI-FI) Initiative, connecting humans and machines via artificial intelligence to enable hypothesis-guided autonomous research. HAI-FI engages all aspects of computing, affording real-time human–machine collaborations that work together to achieve the best path toward advanced knowledge and impactful scientific discovery.
Event Type
Job Posting
TimeWednesday, 16 November 202210am - 3pm CST
Back To Top Button