research · questions before models

Research

What would you do with a forecast for your own backyard?

Mostly I turn data into decisions. I'm finishing a research Master's in Computer Science and tinkering with applied ML side projects.

01

Climate downscaling with hierarchical denoising

research master's · computer science

Global climate models are coarse, typically 25–100 km per grid cell. That's fine for global trends but useless if you need to know what's about to happen over your catchment, vineyard, or grid asset. My research uses a hierarchical denoising approach to take that coarse output and produce high-resolution predictions that match the underlying physics.

Precipitation (var 4) over training: the model's prediction on the left and the residual against ground truth on the right. The prediction develops fine-scale structure as training goes on, while the residual stays small throughout.

Downscaling · Diffusion & denoising · Climate models · Physical consistency

02

Fish-GPT

side project · vision + llm

A toy that wires a vision model into a chat interface, so you can point a camera at a fish (or anything, really) and ask follow-up questions. I built it mainly to see what the failure modes of current vision-language models look like on natural, uncurated input.

Short demo: identification and Q&A on a single fish image.
03

Postcards from the field

Sometimes the science leaks into the writing. Voodoo, Mould, and Models is a postcard from the American Geophysical Union conference in New Orleans.