Machine Learning Research Scientist, Mechanical Intuition in Multimodal Models

The Opportunity

We are looking for a Research Scientist to join us in building intelligent systems for physical assembly. This role is well-suited for a recent PhD graduate with a strong implementation track record and a genuine curiosity about how things are made.

As a researcher on the team, you will design and implement learning pipelines from scratch, run experiments to evaluate a wide range of architectural, data, and algorithmic choices, and help shape how we apply modern machine learning to the challenges of robotic assembly. You will work at the intersection of policy learning, reinforcement learning, and physical reasoning — and have the opportunity to explore how large language models and agentic infrastructure can be brought to bear on real-world manufacturing problems.

Responsibilities