CCI - EVS - Brown, Brian - 11/10/26

Argonne National Laboratory•Lemont, IL

About The Position

Apply the full Adapt radar-analysis workflow to a long-term archive of observations and use the resulting storm features, tracks, and derived products for multi-sensor analysis of convective storms. Develop a transformer- or graph-based nowcasting model using Adapt-derived cell histories, radar observations, and their spatiotemporal evolution to predict short-term cell location and development. The project will demonstrate Adapt as a practical scientific analysis framework through large-scale, reproducible analysis workflows and multi-sensor applications. An important component is systematic error analysis of Adapt outputs, rigorous validation of nowcasting performance, and computational performance. Issues discovered during large-scale application—including failures, edge cases, incorrect feature behavior, performance limitations, or usability problems—will be documented, reproduced, and, where appropriate, addressed through improvements to the Adapt modules.

Requirements

  • The appointment must be conducted entirely within the United States.
  • Applicants must currently be enrolled in undergraduate or graduate studies at an accredited institution, have graduated from an accredited institution within the past 3 months, or be actively enrolled in a graduate program at an accredited institution.
  • Must be at least 18 years old when the appointment begins.
  • Must have a cumulative GPA of at least 3.0 on a 4.0 scale.
  • Must be a U.S. citizen or Legal Permanent Resident at the time of application.
  • Candidates may be required to complete pre-employment drug testing depending on appointment length; students remain subject to applicable drug-testing policies.

Responsibilities

  • Apply the full Adapt radar-analysis workflow to a long-term archive of observations.
  • Use storm features, tracks, and derived products for multi-sensor analysis of convective storms.
  • Develop a transformer- or graph-based nowcasting model using Adapt-derived cell histories, radar observations, and their spatiotemporal evolution.
  • Predict short-term storm-cell location and development.
  • Demonstrate Adapt as a practical scientific analysis framework through large-scale, reproducible analysis workflows and multi-sensor applications.
  • Conduct systematic error analysis of Adapt outputs and rigorous validation of nowcasting performance and computational performance.
  • Document and reproduce issues found during large-scale application and, where appropriate, improve the Adapt modules.
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service