ML Engineer - Power
At a glance
Mid-level Machine Learning Engineering role at Kpler. Paris · full-time.
Pay not stated ML Engineer salary
aimlroles summary, based on the employer's posting.
What you'll do
- Build and run ML pipelines and microservices for power forecasting and electricity-grid models
- Turn statistical and ML experiments into scalable Python applications ready for production
- Shape PostgreSQL data structures and processing for time-series records and event streams
- Automate model training, backtests, evaluation, tuning, and feature and model tracking
- Create dependable data pipelines and contribute tested Python code through reviews and CI/CD
What you bring
- Around two to five years in software engineering focused on data work
- Strong Python experience in large production codebases, beyond notebook-only work
- A solid grasp of grid operations and electricity markets, including generation and transmission
- Hands-on experience with data engineering, databases, ML research, and model and feature versioning
- Working knowledge of Git, code review, Agile methods, and written and spoken English
Who this fits
This mid-level role suits a data-focused engineer who can carry ML work from research into production. It is based in Paris, France, with a hybrid setup.
From the employer
About the Role
As a Machine Learning Engineer at Kpler, you will play a key role in developing and deploying predictive models that power our global commodity, energy, and maritime intelligence platforms. Working closely with Data Scientists, Data Engineers, and Product teams, you will bridge the gap between machine learning experimentation and production-grade software delivery. Your work will directly transform complex data flows into real-time, actionable insights that help world-leading trading firms, industrial leaders, and analysts make critical decisions.
Key Responsibilities
-
Architect and deploy ML pipelines: Design, build, and maintain production-grade machine learning workflows and microservices for power market forecasting and electricity grid modeling.
-
Bridge research and engineering: Transition statistical and machine learning prototypes from initial experimentation into scalable, production-ready Python applications.
-
Manage time-series and event data systems: Design and optimize database schemas in PostgreSQL to handle high-throughput time-series data, event streams, and normalization routines.
-
Implement robust MLOps practices: Establish automated model training, backtesting, evaluation, tuning, and feature/model versioning standards across deployments.
-
Drive data engineering quality: Construct clean ingestion and transformation pipelines, ensuring high integrity, validation, and low-latency access across analytical models.
-
Champion software excellence: Write modular, well-tested Python code, actively participating in peer code reviews, CI/CD automation, and Agile delivery processes.
Experience & Background
-
What you'll need (Must-haves)
-
Software engineering foundation: Approximately two to five years of experience as a data-focused software engineer.
-
Python mastery: Significant experience working with large production Python codebases, rather than working exclusively in notebooks.
-
Domain knowledge: Deep understanding of electricity-grid fundamentals, including generation, transmission, and electricity markets.
-
Data engineering & databases: Experience in data engineering, including working with PostgreSQL or similar databases, database design, data normalisation, and managing time-series and event data.
-
DS & ML research rigor: Proven experience in data science and machine learning research, encompassing statistics, hypothesis testing, model training, evaluation, backtesting, tuning, and model selection.
-
MLOps & versioning: Practical experience in machine learning engineering, specifically including model and feature versioning.
-
Engineering practices & communication: Confidence working with Git, code reviews, and Agile methodologies, supported by strong written and spoken English.
Nice-to-haves
-
Cloud platforms: Experience deploying ML workloads on AWS or GCP using Docker and Kubernetes.
-
Workflow orchestration: Familiarity with orchestration tools such as Apache Airflow, Kubeflow, or MLflow.
-
Streaming technologies: Exposure to real-time streaming architectures (e.g., Apache Kafka).