Machine Learning Engineer

About Us

Over the past few months, with a founding team consisting of Julien, a second-time founder with a PhD in materials science, and Katie, an experienced operator in the climate tech space, Marble has raised its first funding round from leading investors. We secured several letters of intent and demonstrated our capability to boost recycled copper production. Our novel wire-forming technology produces high-conductivity copper wire without requiring 99.95% pure copper raw materials. By eliminating the costly and polluting smelting and refining steps, we reduce the overall production costs while utilizing millions of tons of copper scrap unsuitable for electrical use. Our technology promotes circular, cost-effective copper wire production, essential for the global energy transition. We are now building our core team, and this is a perfect time to join.

About the Role

We are looking for a Machine Learning Engineer with a strong background in deep learning for predictive process control to model and optimize our novel wire forming process. You will lead the development of a model-based controller that anticipates system behavior and drives real-time performance improvements. Your work will bridge physical process modeling, data-driven algorithms, and real-world industrial systems. Contract type is CDI (permanent) and the location is hybrid, based in Nanterre. English fluency is required, and French is a plus but not necessary.

Your Responsibilities

• Develop and deploy predictive control algorithms to optimize mechanical and thermal processes for wire forming.
• Build hybrid models using deep learning and physics-based insights, incorporating process data and literature proxies.
• Design and implement efficient data collection pipelines from characterization methods to support model training and validation, gathering optimized datasets.
• Simulate process dynamics, stress-test control logic, and conduct live tests on physical prototypes.
• Ensure the system is robust to uncertainties and ready for industrial scale-up.

Ideal Background

• Engineering degree, MSc or PhD in Machine Learning, Control Engineering, Applied Mathematics, Physics, Material Science or a related field.
• Strong hands-on experience in model-based control and reinforcement learning for process control/material characterization.
• Proven track record applying deep learning to real-world physical systems.
• Proficiency in Python and ML frameworks (e.g., PyTorch, TensorFlow, C++).
• Familiarity with simulation tools (e.g., Abaqus, Ansys, Simulink, Modelica, or custom simulation environments).
• Bonus: Experience with manufacturing systems or industrial process control. Bonus: Demonstrated initiative in building and experimenting through side projects or hands-on tinkering.

What We Offer

We’re building a world-class industrial deep tech company from the ground up. As one of the earliest team members, you’ll shape the tech, culture, direction, and impact of this start-up. With a CDI contract, competitive compensation, BSPCE package, and 50% health insurance coverage, we also offer a hybrid work model (office in Nanterre). We believe deep tech needs deep diversity. We are committed to building an inclusive, impact-driven team; even if you’re excited about our mission but don’t meet every qualification, we still want to hear from you. We value versatility and the ability to learn, which is essential as we grow. Let’s build the future of circular copper together.