Photo of Jerry Lin

Jerry Lin

Postdoctoral Researcher at Boston University working under Libby Barnes · Boston, MA

work stuff

I am a postdoctoral researcher with Libby Barnes at Boston University. Previously, I completed a PhD studying under Mike Pritchard (now at NVIDIA Corporation) at UC Irvine where I conducted research using machine learning to cheaply emulate computationally expensive subgrid physical processes in multiscale climate models.

I believe machine learning is uniquely suited for positive societal impact in climate science because of access to large volumes of data, an explosion of research and interest in machine learning / AI generally, and the human desire to leave the world in a better place than we found it. If you are a machine learner OR a climate science domain expert looking to collaborate, please do not hesitate to contact me at jlin404@bu.edu.

You can download an up-to-date copy of my resume here.

non work stuff

Outside of immediate research interests, I am broadly interested in the protection of democratic institutions and ways technology can be oriented towards societal progress. I also enjoy running, ebiking, sailing, playing board games, and learning new things. Feel free to send me an email at jlin404@bu.edu if you’d like to chat! 😊

Recent publications

  1. Run It Back: Autoregression vs. Video Diffusion for End-Conditioned Weather Trajectories

    Accepted (peer reviewed), NeurIPS 2026 Workshop on Tackling Climate Change with Machine Learning, 2026

  2. Signoff: Bringing Human Accountability Back to Agent-Powered Research

    Accepted (peer reviewed), Bay Area Machine Learning Symposium (BayLearn), 2026

  3. Extremes on Rewind: Generating 1,000-Member Ensembles Initialized at a Final Condition

    Under review · Preprint, arXiv, 2026

  4. Data-driven integration kernels for interpretable nonlocal operator learning

    Published in Environmental Data Science, 2026

All publications