Robotics Engineer (Physical AI/ VLA Models)

Key Responsibilities:
Develop and implement reinforcement learning (RL), imitation learning, vision-language-action (VLA) models, or behavior cloning policies for physical robots.
Train, test, and validate policies in high-fidelity simulators and transfer them reliably to physical hardware.
Architect and maintain robot software stacks using ROS 2, integrating sensor data (LiDAR, RGB-D cameras, IMUs) and edge compute platforms.
Implement kinematics, dynamics, trajectory optimization, and low-level control systems.
Collaborate with mechanical, electrical, and computer vision teams to debug, benchmark, and stress-test end-to-end robotic applications in real-world lab or field environments.
Build and manage scalable synthetic data generation, domain randomization, and teleoperation data collection pipelines for AI training.
Good to have:
Experience with Vision-Language-Action (VLA) models or foundation models for robotics.
Bachelor’s, Master’s, or Ph.D. in Robotics, Computer Science, Mechanical Engineering, Electrical Engineering, or a related field.
Prior 3+ years of hands-on software and hardware engineering experience in robotics or embodied AI.
Strong proficiency in Python and C++.
Deep experience with ROS / ROS 2, Git, Docker, and CI/CD pipelines.
Experience with PyTorch or JAX, and physics-based simulators.
Solid grasp of forward/inverse kinematics, rigid-body dynamics, state estimation, sensor fusion, and classical/learning-based control.
Proven track record of running and debugging code on real physical robots (e.g., robotic arms, mobile manipulators, humanoids, or quadrupeds).

Skills
Python, C++, Git, Docker, CI/CD, PyTorch, Computer Vision

Experience
3+ yrs

Work arrangement
No