I supervise postdoctoral researchers, graduate students, and undergraduate researchers working at the intersection of atmospheric physics and machine learning, including through the LEAP Momentum Fellows and REU programs. Prospective postdocs and students interested in scientific machine learning for clouds and Earth system models are welcome to get in touch.

---

Current openings

Postdoctoral Research Position: AI, Cloud Microphysics, and Cloud Seeding

We are seeking a postdoctoral researcher to develop AI-enabled approaches for evaluating and optimizing cloud seeding, including whether it could alter precipitation and conditions associated with wildfire risk in the Western United States. The researcher will conduct aerosol-perturbation experiments in three-dimensional large-eddy simulations and develop machine-learning approaches—including generative optimization, reinforcement learning, and reduced-order modeling—to identify effective seeding strategies.

Applicants should have experience in atmospheric science, cloud microphysics, numerical modeling, machine learning, or a closely related field. Experience with large-eddy simulations, particle- or bin-based microphysics, Python, and high-performance computing is particularly desirable. The position offers opportunities to contribute to interdisciplinary research at the intersection of atmospheric modeling, scientific machine learning, precipitation formation, and weather modification. The postdoctoral researcher will be based at Columbia University in the Dept. of Earth and Environmental Engineering and will be co-advised by Dr. Kara Lamb and Prof. Pierre Gentine.

Research Intern: An AI-Ready Benchmark Dataset for Cloud Microphysics

As part of our CM4 DOE Genesis Mission project, we are building a community AI-ready benchmark dataset to systematically evaluate machine-learned and conventional representations of turbulent collision–coalescence, the process through which cloud droplets collide and grow into precipitation. This dataset will draw on an existing library of high-resolution large eddy simulations spanning warm-cloud regimes from meter-scale cloud chambers to cumulus and stratocumulus.

Working with the Columbia team and collaborators at the National Center for Atmospheric Research, the intern will help with preparing the dataset for public release: extracting droplet size distributions and environmental variables into analysis-ready formats, developing metadata, quality control, and reproducible workflows, implementing box-model and bulk microphysics baselines, and producing benchmark metrics. The position offers experience in scientific data engineering, cloud microphysics, reproducible research, and open benchmarks for AI-enabled Earth system modeling.

Eligibility: Columbia students only; MS level preferred.

To apply or request additional information, please send a CV and brief statement of research interests.