Fellow at NRAO

The National Science Foundation (NSF)-Simons CosmicAI Institute seeks to hire a postdoctoral fellow to help initiate a new comprehensive research program to foster foundational advancements in AI through the analysis and processing of astronomical data at the National Radio Astronomy Observatory (NRAO), located on the grounds of the University of Virginia (UVA) in Charlottesville, Virginia. Other NRAO duty stations may be considered if justified.

This work will be carried out in close collaboration with the CosmicAI team members at the University of Utah.

The NSF-Simons AI Institute for Cosmic Origins (CosmicAI) aims to advance transformative AI with the overarching goal of increasing the accessibility of astronomy data and knowledge for researchers, students, and the public. CosmicAI leverages partnerships between academia [UT Austin, Texas Advanced Computing Center (TACC), U. of Virginia, U. of Utah, UCLA], national facilities (NRAO and NSF NOIRLab), nonprofits, and industry to develop capabilities that enable astronomical researchers to conceptualize, define, and execute research projects via trustworthy, efficient, robust, and explainable AI methods.

By building next-generation AI tools, the Institute aims to accelerate discoveries into the most basic human question:
Where do we come from?

The NRAO is a research and development organization that plays a vital role in the study of the universe. The Observatory operates a variety of radio telescopes that span the globe, including the Very Large Array (VLA) in New Mexico, the Very Long Baseline Array (VLBA) across North America, and the Atacama Large Millimeter/submillimeter Array (ALMA) in Chile. These telescopes are among the most advanced in the world, allowing astronomers to explore the universe in unprecedented detail.

Data volumes for interferometric facilities will increase by over two orders of magnitude in the coming decade with instruments such as the Wideband Sensitivity Upgrade (WSU) on Atacama Large Millimeter/submillimeter Array (ALMA), and the next generation Very Large Array (ngVLA). CosmicAI's research efforts at the NRAO will focus on expediting calibration, imaging, and analysis of this high-dimensional data. Although stationed at the NRAO, the CosmicAI fellow will have full access to Institute facilities and equipment (e.g., world-class computational resources at TACC, including Vista, a newly instantiated 600-GPU node cluster) and will routinely collaborate with Institute members across all sites to facilitate the development of cutting-edge AI tools and methodologies.

The CosmicAI fellow at the NRAO will be expected to spend 50% of their time on Observable Universe Working Group research efforts, with the remaining 50% available for independent research; ideally, there will be substantial overlap between the two. The CosmicAI Fellowship appointment will be awarded initially for a two-year period, with an expectation of a third year. Institute-directed research will focus on two main areas:

Developing a methodology and toolset to distinguish good calibrations from problematic calibrations that currently require human evaluation to expedite high-volume, high-throughput calibration and imaging.

Developing robust, automated data segmentation for spatio-kinematic regions of interest in hyperspectral image cubes and leveraging to improve the efficiency of creating compact science-ready data products in imaging pipelines.

As a CosmicAI fellow at NRAO, you will play a leading role in this groundbreaking research, contributing to either or both areas depending on your interest and expertise.

Work is mission-driven, team-oriented, and typically performed in an office setting within a research or development environment.

You have a PhD in Astronomy, Astrophysics, Physics, Data Science, or a related field.

Previous experience in data science, data reduction and analysis, radio astronomy, and/or software development.

Established record of independent research.

Research expertise in interferometric radio/millimeter/submillimeter astronomy.

Proficiency working with a geographically distributed, multi-disciplinary team.

Experience with standard radio astronomy data reduction software (e.g., CASA)

Experience with modern machine learning / deep learning techniques and software packages

Experience with high-performance or high-throughput…