[ES_JOBS_NET] POSTDOCTORAL ASSOCIATE, hyper-local weather forecasting, MIT

[Christine Wiedinmyer]

https://careers.peopleclick.com/careerscp/client_mit/external/jobDetails/jobDetail.html?jobPostId=33971&localeCode=en-us

Posting Description

*POSTDOCTORAL ASSOCIATE, **Mechanical Engineering,* will work under the
direction of Prof. Sherrie Wang to develop deep learning methods for
hyper-local weather forecasting with uncertainty quantification. The
position is supported by a NASA-funded project focused on probabilistic
downscaling of global weather models using satellite remote sensing,
generative AI models, and conformal prediction. The research integrates
numerical weather prediction (NWP) outputs, weather station observations,
and satellite data to generate accurate, uncertainty-aware local forecasts
for applications in disaster response and energy systems. Will develop and
implement machine learning models for local weather forecasting and
uncertainty quantification, including probabilistic and generative
approaches; integrate and analyze heterogeneous datasets, including
numerical weather prediction outputs, weather station observations, and
satellite remote sensing data; design and run experiments to evaluate model
performance and generalization across locations and conditions; contribute
to the preparation of manuscripts, technical reports, and presentations for
scientific and sponsor-facing dissemination; collaborate with project team
members and external partners to align research with application needs in
energy and disaster response; mentor graduate and undergraduate students
and contribute to a collaborative research environment; and participate in
project meetings and related research activities as needed.

Job Requirements

*REQUIRED*: Ph.D. in computer science, electrical engineering, atmospheric
science, or a related field with a strong background in machine learning,
statistical modeling, or geospatial data analysis; experience with deep
learning methods for spatiotemporal data; experience working with
large-scale datasets, ideally including remote sensing or weather/climate
data; strong programming skills in Python and experience with ML
frameworks; and familiarity with uncertainty quantification methods.

4/22/2026
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.ucar.edu/pipermail/es_jobs_net/attachments/20260423/6bc724a2/attachment.htm>