The science behind CervestAddressing some of the most complex and interesting challenges in Earth Science AI and climate.
Our world-leading scientists and researchers have published 60+ peer-reviewed scientific papers in journals ranging from Nature Climate Change and PNAS to Journal of the Royal Statistical Society.
Our key areas of research include: deep learning, non-parametric Bayesian, hybrid modelling, transfer learning, spatio-temporal modelling and multi-task learning.
Current explorations (by research residents) include ethics of climate AI, de-noising clouds though deep learning (GANS), and optimising weather data for dynamic ML modelling.
We are a team of impatient optimists who want to make a lasting, positive and sustainable impact across the planet – by doing science! We believe the benefits of scientific thinking should be accessible, affordable and used to empower those managing nature’s resources – growers, buyers and policymakers – to adapt to rising climate volatility.
As such, we are seeking a diverse set of candidates eager to ask questions that have not been asked before in the fields of earth sciences, AI and statistical sciences, as well as social sciences. A typical project lasts 3-6 months (with flexibility on this).
Past and present Research Residents
Agnes Schim van der Loeff
Research Residents: FAQs
Where can I see some examples of Research Residence projects?
You can learn more about the research projects completed by our residents on our blog. Here you’ll find an example of work completed by Agnes on systemic impacts of combination of AI & Earth sciences at scale. Or Mike’s work on remote sensing denoising using deep latent variable generative models.
What can I expect from the programme?
Our research programme is very similar to an academic Masters or PhD, but with a spin of an open and welcoming start-up culture. You will be reading papers, leading an independent research project, and of course you will receive the best possible mentoring experience from our science and engineering team. Closer to the end of your residency, you will be expected to share your work in the best possible medium, that being a blog post or an academic publication at a conference paper or a workshop.
Who should apply?
Most important for us, is that you are mission driven, hungry to learn, and have a strong interest in earth sciences, machine learning, and the implications of this combinations on policy and societies. Our ideal candidate has either a degree or equivalent experience in STEM fields. We strongly encourage candidates with a non-traditional backgrounds and experiences.