Advance robotic manipulation capabilities, enabling dynamic and physically interactive tasks.
Requirements
- Proficiency in Python, and deep learning frameworks (e.g., PyTorch, JAX).
- Decent understanding of multimodal models, modern ML architectures (transformers, diffusion models, etc.).
- Expertise in imitation learning, reinforcement learning, tactile sensing and robotics learning.
- Proficiency with one or more physical simulators (e.g., MuJoCo, IsaacSim, Drake, PyBullet, PhysX) and experience working in a deployed robotics environment.
- Expertise in neural network deployment (e.g., TensorRT) and GPU programming with CUDA.
- Familiarity with 3D computer vision and/or graphics pipelines
- Experience with Large Language Model.
- Expertise in C++ programming.
Responsibilities
- Design and deploy manipulation algorithms for high-DOF robotic tasks (e.g., grasping, connecting, picking, placing, etc).
- Develop motion planning models for dynamic environments.
- Deploy models as production-ready solutions on RoboForce robots.
- Create and enhance contact-rich robot learning stacks through physics-based simulation.
Other
- Master’s degree in Machine Learning, Robotics, or related field with 4+ years of experience or a PhD degree.
- Requires 5 days/week in-office collaboration with the teams.
- Strong publication on top conferences in robotics manipulation.
