Since about 2008 I've been trying to help reduce CO₂ emissions enough that the climate actually notices. This turns out to be surprisingly tricky!
Between about 2017 to 2023 I applied machine learning to energy forecasting. But weather forecast datasets are hard to work with because they tend to consist of petabytes of legacy file formats (like GRIB)! So, starting in 2024, my new obsession is building tools (in Rust and Python) to help make it easier to work with petabytes of GRIB data in cloud object storage. My current project in this vain is hypergrib.
The "dream" is that anyone with a laptop, an Internet connection, and a little knowledge of machine learning should be able to train and run a state-of-the-art energy forecast (by lazily loading historical and live gridded weather forecasts from multiple providers). We're a long way from this "dream" right now! But we're getting there! Please see this blog post for more info on this idea.
I'm co-founder of Open Climate Fix (a non-profit).
From about 2008 to 2023 to I was quite active on Twitter. Then I dabbled in Mastodon. From late-2024 onwards, my main focus is Bluesky; I'm no longer active on Twitter/X or Mastodon.