Machine Learning Engineer, Power
On-site · Paris (France) · Mid · Full-time
Want offers like this one?
Create your free PyHires profile: you will see how well you match each offer and get a weekly email with the ones that fit you.
Overview
- Work mode
- On-site
- Location
- Paris (France)
- Engagement
- Full-time
- Seniority
- Mid
- Languages
- English
- Open to candidates from
- EU countries only
- Company
- Kpler
- Posted
- Nov 19, 2026
About the role
Kpler provides intelligence on global commodity, energy and maritime markets. As a Machine Learning Engineer you will develop and deploy the predictive models behind its power market forecasting and electricity grid modelling, bridging ML experimentation and production-grade software with Data Scientists, Data Engineers and Product teams.
What you'll do
- Design, build and maintain production ML workflows and microservices for power market forecasting and grid modelling
- Turn statistical and ML prototypes into scalable, production-ready Python applications
- Design and optimise PostgreSQL schemas for high-throughput time-series and event data
- Establish MLOps practices: automated training, backtesting, evaluation, tuning and feature/model versioning
- Build clean ingestion and transformation pipelines with high integrity and low-latency access
- Write modular, well-tested Python and take part in code reviews, CI/CD and agile delivery
What you bring
- About two to five years as a data-focused software engineer
- Significant experience with large production Python codebases, not only notebooks
- Deep understanding of electricity grids: generation, transmission and electricity markets
- Data engineering with PostgreSQL or similar: schema design, normalisation, time-series and event data
- Data science and ML rigour: statistics, hypothesis testing, training, evaluation, backtesting and model selection
- Practical MLOps with model and feature versioning, Git, code reviews and strong English
What they offer
- A global team of 850+ experts from 69 countries
- Work whose output is used by world-leading trading firms and analysts
Requirements
Based on:Python
Must-have
- Git
- Microservices
- PostgreSQLEssential
About Kpler
Kpler simplifies global trade information and turns complex data into insights for clients in commodities, energy and maritime. Founded in 2014, it has more than 850 people from 69 countries.
Stack
- Python
- Git
- Microservices
- PostgreSQL
Company
Kpler
France