About the Role
We are a young startup, looking for a passionate candidate who can contribute to our mission, and create impactful solutions to climate and resilience issues. We are working on the next generation of ML based weather and climate predictions, with an emphasis on cutting edge research and novel technology.
This person’s primary focus will be research and development on subseasonal (3-6 weeks) and seasonal (2-6 months) forecasts. This is a rapidly growing research sector where the combination of dynamical models and machine learning is making big improvements. The role will focus on academic style research on a National Science Foundation funded project using the latest ML algorithms.
We hire people who are collaborative, adaptable, communicate well, and love to learn. Expect to give and receive constructive criticism, as we are constantly seeking to push the innovation frontier while simultaneously growing as individuals and as a team. The position is well suited for someone with strong programming skills and a Ph.D in Atmospheric Sciences, Oceanography, or a related Earth Science discipline. All applicants with relevant experience are encouraged to apply.
- Research and understand the leading academic literature in seasonal and subseasonal forecasting
- Build and implement ML based forecasting algorithms in Python and Tensorflow
- Curate pre-training and operational datasets for training and inference
- Deliver step-change improvements in our industry leading seasonal and subseasonal forecasting
- Share and document your breakthroughs in internal presentations, conference talks, and peer reviewed papers
- PhD or equivalent experience in Atmospheric Science, Oceanography, or a related field
- Experience writing code for data management, analysis, and visualization in Python.
- Experience with machine learning frameworks such as Tensorflow or Pytorch, especially applied to geospatial or time series data
- Experience working on software projects with a team: communication, management, issue tracking, and version control