Robot Learning Engineer

Join Skild AI in our mission to build a scalable, general-purpose brain for a diverse range of robotic form factors. As a Robot Learning Engineer, you will develop foundation models that enable robots to generalize across tasks and environments, moving beyond narrow automation toward true physical intelligence. Your work will directly impact the deployment of autonomous systems in complex, real-world scenarios by bridging the gap between large-scale data and physical actuation.

- Design and train large-scale transformer models for end-to-end robotic manipulation and locomotion. - Develop scalable data pipelines to ingest and process heterogeneous datasets, including human demonstrations and teleoperation data. - Optimize Sim-to-Real transfer techniques to ensure high-fidelity performance on physical hardware. - Implement and experiment with novel architectures for vision-language-action (VLA) models. - Collaborate with the hardware and infrastructure teams to deploy and profile models on edge devices. - Conduct rigorous evaluation of model performance using both automated benchmarks and real-world testing. - Contribute to the company's core codebase to improve the efficiency of our robot learning platform.

- Master’s or PhD in Robotics, Computer Science, or a related field with a focus on Deep Learning. - Proven track record of training large-scale models, particularly in the domains of Reinforcement Learning or Imitation Learning. - Strong proficiency in PyTorch or JAX and experience with distributed training across large GPU clusters. - Experience with physics simulators such as Isaac Gym, MuJoCo, or PyBullet for high-throughput data generation. - Solid understanding of robotic kinematics, dynamics, and low-level control loops. - Experience working with real-world robot hardware and handling the noise and latency inherent in physical systems. - Demonstrated ability to implement and iterate on state-of-the-art research papers in Robot Learning.