Paris | Permanent (CDI) | €75,000–85,000 base plus discretionary bonus

The business

An independent French energy company that develops, owns and operates grid-scale flexible assets and trades them on the power markets. Well funded, small team, and the whole stack built in-house rather than bought in – forecasting through to execution.

The role

You own electricity price forecasting. The models are already live, and they aren’t decision support: they feed the optimisation engine that decides when assets charge, discharge and bid. Forecast error shows up in the P&L the same day.

Short horizons are the priority. Longer-range work exists and can be picked up over time.

What you’ll do

• Maintain and extend the live models, and design new approaches where the current ones fall short
• Build ML and deep learning models alongside fundamental ones
• Develop the scenario generators behind the price views, and put proper uncertainty around the central case
• Turn market fundamentals and asset constraints into usable features
• Work out what drives price and volatility, and quantify it
• Improve the MILP dispatch modules, on economic performance and on speed
• Track forecast error in production and act on it
• Help harden the platform generally

What they’re looking for

• Three to five years forecasting, modelling or quantitative analysis in power markets
• Excellent Python
• Time series and applied ML or deep learning
• Statistics, applied stochastic modelling, mixed-integer optimisation
• A working understanding of how the market functions — day-ahead, intraday, balancing, ancillary services
• Git, code review, CI/CD, and the instinct to take your own work into production
• Master’s or engineering degree in energy, applied maths, data science, econometrics or computer science

Useful but not required: dispatch optimisation, systematic trading, battery storage.

Practicalities

Paris-based. Working language is French with English spoken alongside it; French preferred rather than mandatory.

Why people take it

Wide scope and real ownership in a small team. Models running against live assets rather than sitting in studies. Close proximity to how the business makes its capital decisions.

Contact: Hamish Graham – [email protected]