We're back for 2023 🌍
We are an initiative uniting the student AI communities of 18 universities across the UK, US and Canada to help reduce carbon emissions by up to 100 kilotonnes per year with machine learning.
Join our Discord server to get involved.
18 world-leading universities
Bristol. Caltech. Carnegie Mellon University. Harvard University. Illinois Urbana-Champaign. Imperial. KCL. Manchester. Michigan. Oxford. Princeton. St Andrews. Toronto. UC Berkeley. UCLA. UCL. Warwick. Waterloo.
We are challenging participants to advance the state of the art in site-level solar power forecasting using satellite imagery, weather forecasts and aerosol data.
Your cutting-edge machine learning contributions could help save up to 100 kilotonnes in carbon emissons each year in Great Britain alone by supporting the research work of the non-profit lab Open Climate Fix for the National Grid Electricity System Operator.
Find out more on the DOXA AI platform.
Participants compete individually, and the top three entrants from each university will be invited to form a team for the finals.
After the qualifying round ends, teams develop their final models and present their solution at finals hosted at UCL and Harvard University.
Participants must be enrolled as an undergraduate, graduate or PhD student at one of the eligible universities in order to take part. Submissions are individual in the qualifying round, but we encourage collaboration: we are one team working to tackle climate change.
Concurrent Finals in
London and Boston
The competition will conclude with simultaneous in-person finals hosted at UCL and Harvard University, bringing together the top three participants from each university at the end of the qualifying round.
Hotels for finalists will be fully paid for and travel in North America will be subsidised.
The finals are an opportunity to present their work to our panel of expert judges, who will then decide the winners of the competition.
Prize Pool of £20,000 / ~$25,000
Split amongst the finalists of each winning team