Big news!! 🚀 Aurora, a foundation model for the Earth system, has been published in @Nature. A massive congrats to the whole team! Very proud of this achievement, and thrilled to see that it’s finally out there. 😊.

Aurora is fully open! 🥳. The air pollution model 🌬️, the ocean wave model 🌊, and the TC tracker 🌀 are now available. And that's not all: all model weights (pretrained and fine-tuned) are now released under the MIT license. 😎. GitHub: #AIforGood.

Wondering how you can forecast whether you can get barrelled tomorrow or not? Tomorrow I’ll be in London to talk about Aurora at UCL, so come check it out! :). (Credits to Magnus Ross for the great promo poster.)

Exciting job update! I am now with the Alan Turing Institute (@turinginst) in London. Our team’s goal is to advance the SOTA in environmental forecasting using end-to-end ML :). Do reach out if you’d like to chat.

RT @peteratmsr: We're excited to reveal this major advance in computational chemistry, an AI model called Skala @MSFTResearch for the excha….

RT @vdbergrianne: 🚀 After two+ years of intense research, we’re thrilled to introduce Skala — a scalable deep learning density functional t….

RT @MSFTResearch: Explore the future of environmental forecasting with Aurora, an AI model from Microsoft Research. In this Abstracts episo….

RT @SilurianAI: 🌍Aurora—now published in Nature—is the first AI to unify weather, air quality, wave & hurricane simulations, while outperfo….

RT @satyanadella: Aurora is a new foundation model from @MSFTResearch that goes beyond weather forecasting, delivering faster, more accurat….

RT @peteratmsr: Just published in @Nature, AI foundation model Aurora @MSFTResearch. It gives the most accurate predictions, not only of we….

Incredibly excited to see that Aardvark-Weather is finally out in Nature!! An amazing project with a truly fantastic team. The lead authors Anna Allen and @stratosmarkou have worked really really hard to make this project happen.

`fasttransform` has just been released, and it's powered by Plum (`plum-dispatch`)! 🚀 Very exciting! Check out their release post, and check out Plum on GitHub to see what Plum can do for you.

RT @ParisPerdikaris: 🎉Excited to announce major new breakthroughs in our Aurora foundation model! Our team (@crisbodnar, @ikwess, @a_lucic,….

I am now on 🦋 too! Find me there under the handle :).

Yesterday I spoke about foundation models for Earth systems at the @lorentzcenter workshop for advancing ecosystem carbon flux research. It is important to connect with those that could potentially use your research, to see what the current needs and problems really are.

Really excited about this! We've put effort into making the model accessible with code that fits in a Tweet. 😊. pip install microsoft-aurora. from aurora import Aurora. model = Aurora().model.load_checkpoint("microsoft/aurora", "aurora-0.25-finetuned.ckpt").

Finally, I'm very pleased that my thesis should compile without any LaTeX warnings. 😎 It turns out that LaTeX warnings convey useful information, and resolving them is in your interest! Who would’ve thought?!. 🧵 8/8.

Throughout the PhD, my supervisor Rich Turner has been nothing short of absolutely amazing. Go work with him if you can!! Unfortunately Rich is not on X, but I wanted to give him a shout-out anyway. 🧵 7/8.

The most fun chapter is probably Chapter 7, discussing the design of the Python package `neuralprocesses`. This chapter is a little unconventional, and will probably never be seriously published. Nevertheless, it was super fun to write, and my examiners really liked it. 🧵 6/8.

However, I don’t think one should view the thesis as one’s life’s work! IMO, ultimately a PhD is not about producing. Instead, it is primarily an opportunity to further develop yourself. You will have enough time to rack up citations afterwards. 🧵 5/8.