Data Scientist (AI)
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Описание вакансии
TL;DR
Data Scientist (AI): Improving high-accuracy wind and solar forecasting models with an accent on time-series analysis, geospatial data, and production model deployment. Focus on building robust ML pipelines, diagnosing production behavior, and ensuring reproducibility in a mission-driven environment.
Location: Hybrid working with offices in Tallinn and Rome; preference is given to candidates based in Tallinn or Rome.
Company
develops high-accuracy wind and solar forecasting using advanced machine learning and weather modeling.
What you will do
- Contribute to forecasting model development, including experimentation and feature engineering.
- Clean, analyze, and transform complex meteorological and time-series datasets.
- Support model deployments through scheduled inference, retraining workflows, and performance monitoring.
- Investigate data and model issues in production to drive effective resolutions.
- Collaborate with the team on modeling directions, documentation, and structured workflows.
Requirements
- 3-5 years of end-to-end ML experience (data to deployment).
- Solid Python skills and experience with PyTorch or comparable deep learning frameworks.
- Intuition for time-series modeling or spatiotemporal data.
- Experience deploying models into production environments.
- Ability to work in a hybrid setup, preferably based in Tallinn or Rome.
Nice to have
- Exposure to weather, NWP, or other scientific datasets.
- Experience with large-array formats like Zarr or Icechunk.
- Familiarity with model monitoring, drift detection, and ensemble forecasting.
Culture & Benefits
- High autonomy and clear ownership over the forecasting product.
- Opportunities to collaborate with experienced data scientists and MLOps engineers.
- Direct impact on clean energy adoption and the transition to net-zero.
- Growth-oriented, transparent environment with high visibility to leadership.
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 →