aipythonmachine learningpytorchremote sensingjaxuncertainty quantificationfinite element modelingice dynamics
Вакансия из Hirify Global, списка международных tech-компаний Для мэтча и отклика нужен Plus
Пожаловаться
Зарплата и рынок
Очень мало
$120к-150к/год
$210к$250кмедиана$310к
По похожим ролям (50) ~ $250к/год
На СНГ рынке: $4.1к/мес ($3.6к - $5.3к)
88
Крутая вакансия
развернуть
Высокозначимая роль на стыке AI и физики для борьбы с изменением климата. Оффер очень прозрачный, с отличным соцпакетом, хотя зарплата ближе к некоммерческому сектору, чем к топовым AI-лабораториям Кремниевой долины.
Research Scientist (AI) (Cryospheric Modeling): Developing data-driven physics and machine learning methods to forecast glacier ice flow and sea-level rise with an accent on remote sensing, probabilistic modeling, and large-scale ice dynamics simulations. Focus on integrating physical domain knowledge, quantifying uncertainty in noisy geoscientific data, and building scalable finite element modeling workflows.
Location: Fully remote with occasional in-person meetings as needed.
Salary: $120,000–$150,000 annually, commensurate with experience.
Company
Arête Glacier Initiative is a newly launched nonprofit initiative fiscally sponsored by hirify.global, focused on understanding Antarctic ice-sheet risks and potential interventions to mitigate sea-level rise.
What you will do
Develop methods to forecast glacier ice flow and its impact on sea-level rise with data scientists, engineers, and scientists.
Design, adapt, and evaluate probabilistic machine learning and hybrid physics–ML models for ice-sheet flow, surface mass balance, and ocean–ice interactions.
Develop and run large-scale numerical simulations of ice dynamics and scalable finite element modeling workflows.
Process and curate satellite, radar sounding, ice-sheet model, and synthetic datasets for model training and validation.
Integrate domain knowledge into model design, interpret outputs under physically realistic conditions, and analyze sensitivity to imperfect or incomplete data.
Participate in regular research reviews and contribute to cross-functional, multidisciplinary collaboration.
Requirements
PhD in atmospheric science, geophysics, applied mathematics, computer science, or a closely related field.
3+ years of postdoctoral or equivalent research experience applying machine learning or deep learning to weather forecasting or Earth system modeling.
Proficiency in Python and PyTorch or JAX, with experience in high-performance or cloud computing and large-scale parallel processing.
Experience with noisy geoscientific data, uncertainty propagation, error estimation, and remote sensing datasets including satellite radar, altimetry, and optical imagery.
Ability to communicate complex technical concepts clearly in written and spoken English and work independently in a remote, fast-paced environment.
Availability for occasional in-person meetings is required.
Nice to have
Experience with ISSM, PISM, MALI, physics-informed neural networks, neural operators, graph neural networks, or AI-accelerated finite element modeling.
Knowledge of Bayesian inference, ensembles, sensitivity analysis, or adjoint methods.
Experience with Zarr, xarray, and Dask for large gridded datasets.
Culture & Benefits
Mission-driven nonprofit environment focused on Antarctic science, climate risk, and informed societal decisions.
Flexible, fully remote work with occasional in-person meetings.
Health, dental, and vision insurance with employee premiums fully covered and 85% of dependent premiums covered.
HSA with $275 monthly employer contributions, FSA, wellness reimbursement, and disability and life insurance.
401(k) plan with 6% matching, 15 paid vacation days, 16 paid holidays, bereavement leave, and paid parental leave.
Hiring process
Submit a resume and a cover letter outlining qualifications.
Interviews are conducted virtually.
Будьте осторожны: если работодатель просит войти в их систему, используя iCloud/Google, прислать код/пароль, запустить код/ПО, не делайте этого - это мошенники. Обязательно жмите "Пожаловаться" или пишите в поддержку. Подробнее в гайде →