Junior ML Engineer — Clean Energy Forecasting

Valentina Cruz

Full-time · Junior · Berlin

This role is no longer accepting applications

Similar open roles you can apply to now:

About the role

We build software that helps energy grid operators and electricity traders manage the variability introduced by renewable generation. Our forecasting models predict solar and wind output at 15-minute intervals for assets across 12 European markets. Accurate forecasts reduce balancing costs, which reduces the economic friction of integrating renewable energy, which is the problem we exist to solve. We have 22 people. We have been profitable since year three. We are not a research organisation — we are an engineering organisation that solves a real operational problem for paying customers who compare our forecast accuracy against our competitors every day. We are hiring our second Junior ML Engineer. The first one joined two years ago as a recent graduate and is now doing excellent independent work. We think that tells you something about how we approach junior development: we invest in it seriously, we pair new engineers with experienced ones, and we are patient about the growth curve in exchange for genuine curiosity and the willingness to do the work carefully. The role is in our core modelling team. You will work on weather data processing, feature engineering from NWP (numerical weather prediction) outputs, and the evaluation pipelines that measure forecast accuracy against actual generation.

Responsibilities

Process and validate NWP weather data feeds from multiple providers for use in our forecasting pipeline
Build and maintain feature engineering pipelines that transform raw weather model outputs into model-ready inputs
Contribute to forecast evaluation reporting — calculating accuracy metrics, diagnosing systematic errors, and summarising results clearly
Support senior engineers in model retraining and validation workflows
Document your work clearly — methodology notes, data quality observations, and pipeline changes

Requirements

A degree in a quantitative discipline — physics, mathematics, computer science, meteorology, or similar — or equivalent self-taught background with demonstrable projects
Python for data analysis and modelling — you are comfortable with NumPy, Pandas, and basic scripting in a shared codebase
SQL for querying generation and weather data from our data warehouse
Foundational statistics: distributions, regression, and time-series concepts you can explain without looking them up
Scikit-learn for supervised learning tasks under senior guidance — you understand what the algorithms are doing, not just how to call them
Curiosity about physical systems, weather, or energy is genuinely useful in this domain
Any experience with meteorological data, NWP outputs, or energy systems is welcome but not expected

Benefits

$56,000 – $71,000 base salary
Full remote — our team spans Germany, France, Netherlands, UK, and Spain
Dedicated mentorship from a senior ML engineer for the first twelve months
$800 annual learning budget
Work on a problem where forecast accuracy translates directly into the economics of the clean energy transition