Data Scientist for Quantitative Modelling

Salary: 55.000 - 80.000 € per year Requirements: MSc or PhD in Data Science, Statistics, Mathematics, Physics, Computer Science, Operations Research, or a related quantitative field. 10+ years of professional experience applying data science or quantitative modelling in an energy trading, energy tech, or commodity trading environment. Proven track record of developing and deploying IT/data-driven solutions that directly support trading decisions or automated dispatch using MLOps tooling and CI/CD for model deployment. Deep understanding of European electricity markets (EPEX, Nord Pool, or equivalent) including day-ahead, intraday continuous, and balancing mechanisms. Excellent programming skills in Python (pandas, NumPy, scikit-learn, LightGBM/XGBoost, or similar); SQL and cloud-based data platforms. Experience with reinforcement learning, Bayesian methods, or time-series deep learning (LSTMs, Transformers) in a trading context. Strong experience with optimisation techniques (LP/MILP, stochastic optimisation) applied to asset scheduling or portfolio optimisation. Excellent communication skills: you can explain a complex model to a trader at 7 AM and defend your methodology in a technical review at 3 PM. Autonomous and self-driven: you take ownership of your roadmap items, push them forward without constant guidance, and know when to escalate. Strong team player: you thrive in a fast-paced, collaborative environment where traders, developers, and data scientists sit side by side. Responsibilities: Take ownership of the development and optimisation of quantitative trading strategies for dispatching flexible assets across intraday, day-ahead, balancing, and ancillary service markets. Build in collaboration with other teams predictive models for price forecasting, asset availability, imbalance signals, and market spread identification to improve bidding and scheduling decisions. Own the trading strategy roadmap, identify, prioritise, and deliver new features and model improvements in iterative cycles aligned with business value. Collaborate closely with traders to validate hypotheses, back-test strategies against real P&L, and incorporate trader intuition into model design. Scale strategies across geographies and asset classes, adapting to local market rules, grid codes, and asset-specific technical constraints (e.g., degradation, ramp rates, state-of-charge). Design and maintain robust data pipelines that feed real-time and historical market, weather, and asset data into modelling and decision engines. Monitor live strategy performance, detect drift or anomalies, and implement rapid feedback loops for continuous improvement. Communicate results clearly to both technical and non-technical stakeholders, translate complex model outputs into trading insights and strategic recommendations. Stay current with developments, and state-of-the-art methods in ML/optimisation relevant to energy trading. Technologies: CI/CD Cloud Support Kubernetes Python SQL numpy pandas More: We are Ironwood Digital, a subsidiary of the E.ON Group based in Essen. We coordinate access to trading markets for our regional business units, bundle opportunities and risks, and provide innovative services. Our core competences include portfolio strategies, risk management, and data processing, and we operate across several European countries as we help shape the future of energy. We are looking for a motivated and curious Working Student to join our dynamic team in Essen as Data Scientist - Flex Trading Strategy Development (f/m/d), with hands-on experience in modern technologies such as Kubernetes and testing frameworks. We offer a flexible hybrid work model, flat hierarchies, opportunities for career entry and personal growth, a modern work environment, health and nutrition benefits, and a central location with excellent public transport connections, free parking, and e-vehicle charging points. We value diversity, inclusion, sustainability, teamwork, appreciation, work-life balance, health, networking, development, and onboarding. last updated 27 week of 2026