Weather and Climate Data Scientist

Shell plc is seeking a Weather and Climate Data Scientist in Bengaluru, India, to join their Data Science team. The role involves developing data science solutions for weather and climate applications, focusing on forecasting and data management. Candidates should have a strong background in relevant fields and experience with machine learning and data analysis. There is an option to work remotely or co-locate with the team in London.
statistical We have an exciting opportunity for a Weather and Climate Data Scientist to join Shell’s Data Science organization. Be part of a team focused on developing end-to-end process maintenance solutions and contribute to delivering value for our customers, and for Shell.

Where you fit

Shell is a global group of energy and petrochemicals companies, with approximately 90,000 employees in more than 80 countries and territories. Shell helps to meet the world's growing demand for energy in economically, environmentally, and socially responsible ways.

In this role, you will be part of the Industrial Data Science team for weather and climate applications for Renewable Energy Solutions within the Data Science Centre of Excellence at Shell. You will join an existing team working on a forecasting engine dealing with structured and un-structured data. You have an option to either work virtually, or co-located with the team in London, working closely with our internal customers, using the lean start-up approach and ScrumBan to accelerate the development of the solution.

What’s the role?

As Weather and Climate Data Scientist, you will be responsible for developing and delivering data science solutions in an agile fashion to the business, to provide insights & improvement to business processes, or support the creation of innovative new business domains. You are expected to have a good understanding of both data-driven empirical models and process-driven physical models. Ability to work with large physical models representing atmospheric processes written in Fortran code, to understand, modify and predict complex natural phenomena, using the state-of-the-art tools are essential to the role.

In addition, you will play a crucial role in designing and building weather and climate database to support business decision making. You will manage large amounts of data; merge data sources; ensure consistency of datasets; create visualizations to aid in understanding data; build machine learning models using the data; and present and communicate the data insights/findings and produce and present results with dashboards.

Further Responsibilities are as follows:
• Researching machine learning models for forecasting and time series analysis using the latest methods
• Write clean and maintainable production-level code, including tests; the tech stack we work with includes (you don’t need to have an experience with all of them): Fortran, Python (Pandas, Scikit Learn, etc.), GIT, Docker, Azure, SQL.
• Work closely with the customer and the Product Owner day-to-day
• Work in a highly-collaborative, friendly Agile environment, participate in Ceremonies and Continuous Improvement activities.
• Documenting and explaining the results of analysis or modelling to both a technical and non-technical audience.
• Learning new engineering practices, technologies and continuously improving our Agile practices.

What we need from you
• Minimum of MSc in Meteorological or master’s degree in any relevant discipline such as Maths, Physics, Statistics and Computer Science.
• Minimum of 5 to 12 years of overall work experience with minimum of 3 years relevant experience working with weather/climate forecast models or applying AI technologies to model predictions to improve forecast quality.
• A good understanding of mathematical modelling of the physical processes. Optimization methods constrained by economic and operational parameters.
• A good working knowledge of large Fortran code bases on cloud computing platforms such as Azure or AWS. Experience in recent versions of Fortran required.
• Good understanding of model output post-processing techniques (e.g., Ensembles, Model Averaging, Model Selection) to deliver forecasts for Renewable Energy Solutions using Python or R.
• Application of Deep Learning Techniques for Renewable Power Forecasting applications, using Timeseries (e.g., LSTMs, GANs) and other unstructured data.
• Ability to write clean, elegant, and maintainable production-level code in Fortran and Python. Experience in version control, testing and refactoring the code.
• Experience with the Python scientific stack, e.g., NumPy, Pandas, Scikit-learn, PyMc3 or similar, etc.
• Experience in weather modelling and renewable forecasting, using opensource code base such as WRF.
• 3D/4D data assimilation techniques used in physical models to improve forecast quality.
• Good understanding of the role of weather and climate on renewable energy generation, storage and transmission