Research Engineer, Climate Modeling

What you'll do

• Adapt NeuralGCM’s training pipeline and data requirements for efficient use on Google Cloud and NVIDIA GPUs, and document and distribute the code.
• Implement and test coupling NeuralGCM with an ML-based ocean model and develop approaches to improve performance in unseen climates.
• Collaborate on open-source development and present tutorials to climate-modeling user groups.

What they ask for

• 2–4 years building and deploying a relevant ML application in a practical or academic research setting; fluency in Python and collaborative coding practices.
• Experience developing ML training code for cloud computing environments.
• B.Sc. in computer or computational science, atmospheric science or a related geophysical science, or applied mathematics/statistics.
• Excellent communication skills.

Nice to have

• Ph.D. in a listed field, with at least one peer-reviewed first-authored publication or accepted conference paper extensively using ML in a physical science setting.
• Training or practical experience in physically based model development, or experience writing ML code in JAX.

Pay

$146,880 - $220,320 USD (base salary)

Tools

Python, NeuralGCM, Google Cloud, NVIDIA GPUs, JAX, TPUs, GitHub

AI focus

Develop an AI-first, physics-informed hybrid climate model that improves multidecadal forecasts and generalizes to unseen climates; couple it with an ML-based ocean model.