Senior Data Scientist

Companion.energy; Oost-Vlaanderen; Hybrid; EUR 70,000 - 110,000; Full time

Job summary

Companion.energy in Ghent, Belgium (hybrid) is hiring a Senior Data Scientist to develop forecasting, optimization, and control algorithms at the core of our platform. You will design time-series models for energy demand and market prices, build real-time dispatch and optimization pipelines, and validate performance with backtests and live measurements in a fast-scaling energy tech environment.

You'll own the data path from raw signals to deployed models, collaborating with engineers and

Qualifications

• Deep experience with time series forecasting on messy, real-world data.

Responsibilities

• Develop forecasting and market price prediction models.

Skills

Time series forecasting

Statistical modeling

Python production

End-to-end data pipelines

Communication with engineers

Tools

Dagster

Airflow

Prefect

PyTorch

Pandas

NumPy/SciPy

scikit-learn

SQL

Job description

TL;DR

We help large enterprises cut energy costs and valorize flexibility by steering their energy use in sync with market prices and renewable generation.

We're hiring a Senior Data Scientist to develop the forecasting, optimization, and control algorithms at the core of Companion's platform.

You'll work on energy demand forecasting, market price prediction, and real-time asset control: models and algorithms that drive second-by-second decisions worth millions.

About Companion.energy

Companion.energy connects the financial side of energy management (contracts, markets, risk) with the operational side (assets, processes, sites). We model complex energy contracts and flexible assets, forecast demand and production, and translate predictions into automated, second-by-second control decisions that move megawatts and money. Our software is used by large B2B enterprises and energy players to lower OPEX, maximize revenue, increase renewable usage, and manage risk.

Why Companion.energy?

• Work on hard problems: Real-time forecasting, optimization, and closed-loop control in complex, stochastic energy systems. Your models and controllers move megawatts.


• See your work in production: Models you build get deployed and measured against real financial outcomes daily. Controllers you design steer real assets in real time.


• Join early, help shape the future: Small team, big impact. You'll have a large say in how we forecast, optimize, and control.


• A mission that matters: Better forecasting and smarter control accelerate the transition to a flexible, renewable grid.

What you'll work on

• Forecasting models for energy demand, renewable production, and market prices — including the uncertainty estimates that feed downstream control.


• Control algorithms that translate forecasts and market signals into real-time asset dispatch decisions (battery steering, load shifting, balancing market participation), accounting for the inherent stochasticity of prices, generation, and demand.


• Optimization pipelines that schedule flexibility across portfolios of assets and markets under uncertainty.


• The data path that feeds all of the above: ingesting and cleaning messy asset and market time series, handling gaps and backfills, quality checks, and the orchestrated pipelines (we run Dagster) that keep models fed and predictions flowing on schedule.


• Closing the loop: connecting prediction, planning, and execution into systems that operate autonomously in a stochastic environment. Concretely, this includes building the backtesting and measurement harness that tells us whether a model is earning money in production — and keeps telling us once it starts to drift.

To be explicit about what this means at our size: there is no data team behind you and no MLOps engineer to hand a model to. You own the path from raw data to a running, monitored model. That is real ownership over something that matters, and it is not a fit for someone who wants to work only in notebooks.

What we are looking for

• Deep experience with time series forecasting and statistical modeling on messy, real-world data.


• Proficiency in Python and the standard data/scientific computing stack (pandas, NumPy/SciPy, scikit-learn, PyTorch or equivalent), at a level where your code goes to production and stays there.


• Models you have shipped, measured against real outcomes, and later had to change.


• Comfortable owning the data path end to end: SQL, and orchestration with Dagster, Airflow, Prefect or equivalent, including the discipline to keep scheduled pipelines healthy rather than just running.


• You can explain a modeling decision to an engineer who doesn't share your maths background, and write down your assumptions clearly enough that someone else can safely change your model.


• Pragmatic approach: you understand the difference between a theoretically optimal solution and one that ships and delivers value under real-world uncertainty.

Strong plusses

• Control algorithms for real-time decision-making under uncertainty and constraints: model predictive control (MPC), stochastic optimal control, or similar.


• Mathematical optimization: linear/quadratic programming, mixed-integer formulations, stochastic programming.


• Knowledge of energy systems: electricity markets, load forecasting, asset dispatch, balancing mechanisms, or related domains.

Location

Ghent, Belgium (hybrid)

How we hire

Intro call → technical case → follow-up conversations with the team and founders.

We keep it simple and can move fast.