Postdoctoral Researcher - AI for Earth System Sciences

Argonne National Laboratory • Lemont, IL

6d • Onsite

About The Position

Argonne National Laboratory has an opening for a highly motivated post-doctoral appointee at the Decision and Infrastructure Science (DIS) Division. The laboratory has developed AI foundation models for weather, seasonal-to-subseasonal forecasting, hydrology, and urban digital twins. This role will involve collaborating with scientists to further develop, benchmark, and deploy these AI foundation models for applications in infrastructure and resilience workflows, supporting both DOE and private sector utilities. The position is focused on applying and developing machine learning-based weather models for the S2S timeframe, with an emphasis on generative AI techniques, evaluating these models, and contributing to research aimed at advancing predictability.

Requirements

• Recent or soon-to-be-completed PhD (completed within the last 0-5 years) in geophysical sciences, computer science, or machine learning.
• 0 to 2 years of experience.
• Knowledge of deep learning, PyTorch/JAX, and scaling deep learning models to large GPU-based machines.
• Knowledge in training domain-specific AI foundation models built using transformers, GNN, and diffusion.
• Technical knowledge in using HPC systems for visualization and analysis.
• Technical knowledge of large, dynamical systems (preferably the atmosphere).
• Knowledge and experience in writing scientific code.
• Skills in clear, concise writing of technical papers.
• Problem-solving skills.
• Organizational skills and flexibility in coordinating a broad spectrum of activities.
• Knowledge of data analysis.
• Ability to work and communicate with stakeholders from public and private sectors.

Nice To Haves

• Experience in developing and verifying deep learning-based models for large dynamical systems (e.g., weather, hydrology).
• Expertise in data and model parallelisms for distributed training on large GPU-based machines.
• Experience using diffusion-based or other generative AI methods and multimodal data training approaches, fine-tuning, and optimizing inference.
• Background in atmospheric science, especially weather modeling.
• Knowledge of atmospheric dynamics, process scale models, and numerical computation techniques.
• Knowledge of using atmospheric and hydrological observational datasets, data assimilation techniques, and statistics.
• Some familiarity in subseasonal-to-seasonal modeling and/or coupled atmosphere-ocean modeling.

Responsibilities

• Collaborate with scientists in DIS, Environmental Sciences division, and Argonne Leadership Computing Facility to further develop AI foundation models.
• Benchmark and deploy AI foundation models for applications.
• Contribute technical expertise through analysis and support for programs and projects related to machine learning, HPC, and computational problems in earth system science and other dynamical systems.
• Develop, evaluate, and apply machine learning/computational approaches.
• Engage in synthesis activities, develop computational tools, compile results, and prepare reports, publications, and documentation.
• Apply and develop machine learning-based weather models for the S2S timeframe with an emphasis on generative AI techniques.
• Evaluate machine learning-based weather models.
• Work with a team of scientists interested in pushing the boundary of predictability.
• Interact and communicate effectively with colleagues.
• Coordinate a broad spectrum of activities.

Benefits

• Comprehensive benefits are part of the total rewards package.

What This Job Offers

• Job Type: Full-time
• Career Level: Entry Level
• Education Level: Ph.D. or professional degree