Domain Data Scientist (Digitization System Operation) (w/m/d)

We are 50Hertz – one of the four transmission system operators in Germany. Together with the Belgian electricity grid operator Elia (Brussels), we form the internationally active Elia Group. At 50Hertz, over 2,100 people work to ensure that current flows reliably, low-loss and optimally controlled at all times over long distances. As a TSO, we play a central role in the implementation of the energy transition in a social, technical and economic sense. Our goal: By 2032 we want to integrate 100 percent renewable energy safely and affordable into the grid and system. Innovative We are driving the energy transition with future technologies and creative solutions. We are already thinking about this day and are ready to go new ways. Divers We work with people from different cultures and with diverse talents to achieve great things together. We always see diversity in every dimension as a chance. Responsible We are responsible for a reliable power supply and the sustainable expansion of the power grid. In order to master this task, it is self-evident to assume responsibility for each other. Community Our strong community and valuable cooperation ensure the success of our mission. This is why we actively promote the individual strengths and interests of our employees and support personal development.

Domain Data Scientist (Digitization System Operation) (w/m/d)

Full time (37 hours per week) Heidestraße 2, 10557 Berlin Unlimited To the next possible date Job ID: 10395

That makes this job interesting for me: I want to be an active part of the energy transition, to design the digital future of the overtime network operators with the team and to develop the Next Generation Control Center Ecosystem (Modular Control Center System)! I take responsibility as a Domain Data Scientist (Digitization System Operation) and, together with other data specialists, provide for high-performance forecast models, consistent network calculations and correct modeling of the system. These include the identification and analysis of use cases with the business stakeholders, their implementation, operationalization and monitoring of software products. I help develop a uniform data strategy and implement new machine learning services for a secure power grid of tomorrow. Together with the team, I will make the highly complex data controllable and predictable within our transmission network to reach “100 percent by 2032: Repayable energy for a strong economy” (https://www.50hertz.com/en/Company/strategy).

With modern working environments at the locations in Berlin and Neuenhagen and an attractive solution for mobile work, 50Hertz offers ideal conditions for me and my personal development. My tasks I am responsible for the new and further development and integration of forecasting and network calculation applications (e.g. feed forecasts, load forecasts, scenario analysis, topology optimizations) using data science methods, I develop machine learning models in the field of renewable energies (end-to-end): from the development of use cases with business experts and product managers to data analysis to the operation of machine learning models in productive use, I collect, clean and transform time series and weather data to ensure high-quality data sets for forecasting and network models and thus to ensure our core processes and follow-up processes (publications, billing), I use modern modeling approaches (e.g. regression models, gradient boosting, deep learning, probabilistic models) to enable precise predictions for energy generation and consumption, I care for clear and comprehensible documentation of code, data pipelines and models for long-term maintenance and further development, I work closely with developers, data engineers, data architects, and users within the system management system, I work with national and international working groups and scientific and technical projects to transfer the best solutions to our operational processes.

I have several years of professional experience in data analytics, data science and machine learning, in particular in the development and operation of data-based solutions, I have a well-founded knowledge of building data pipelines, data handling and handling large volumes of data as well as relational and NoSQL databases (e.g. MongoDB), I have very good programming knowledge in Python including common frameworks (e.g. Pandas, Scikit-learn, TensorFlow) as well as experience with API development, deployment and monitoring of ML services, I bring several years of experience in the application and evaluation of ML algorithms, feature engineering and model evaluation (e.g. cross-validation, metrics such as RMSE, MAPE), I know myself in the field of monitoring and observability of ML services (e.g. with Prometheus, Grafana) as well as experience with deployment and scaling of models in production environments, I'm sure about Git and modern development workflows (Code Reviews, CI/CD). I am characterized by pronounced analytical thinking, self-initiative and the ability to explain complex technical issues in a manner that is appropriate to the address, I have fluent German knowledge (C1) and very good English knowledge (B2) in word and writing. No need, but of advantage knowledge of the development and productive use of data models in the environment of the energy industry, for example for forecasting, network calculation and optimization models, experience in dealing with the Common Information Model (CIM) in the Common Grid Model Exchange Standard (CGMES) and UCTE data formats, Experience with cloud services such as AWS SageMaker, Google Cloud AI or Azure ML, experience with ELT/ETL frameworks, CI/CD, deployment in the Kubernet environment with Azure, Docker Containers, streaming architectures and technologies such as Apache Kafka, Ideally experience in high-performance computing and development of solution algorithms, Ideally experience with load flow calculations and network analysis in electrical networks with the help of, for example, PowerFactory, Pandapower, Integral or Neplan.

Chance to advance the energy transition itself Pronounced collegial environment Various working time models with long-term account Promoting personal development 30 days off Employee events (e.g. team development days and company festival) Fair remuneration and extensive social benefits Professional consultancy on (psychical) health, care, law, finance and more Occupational pension Flexikompass – this is so flexible: Working time: 37 hours per week (full time), Flexible working hours with long-term account to support your work-life balance, The number of sliding days is not limited, No core working time, Mobile work (also from home) possible. .