Robotics Autonomy Engineer - Manipulation

Job description

FieldAI’s Irvine team is where embodied AI meets real robots, real sensors, and real field deployments. Based in the heart of Southern California’s robotics ecosystem, we build risk-aware, reliable, field-ready AI systems that solve the hardest problems in robotics and unlock the full potential of embodied intelligence. If you want your work to ship, get tested on hardware, and improve through real deployments, Irvine is the place. We go beyond typical data-driven approaches or pure transformer-only architectures, combining rigorous engineering with learning systems proven in globally deployed solutions that deliver results today and get better every time our robots run in the field. About the Job Field AI is building the future of autonomy—from rugged terrain to real-world deployment. We’re on a mission to develop intelligent, adaptable robotic systems that operate beyond simulation and thrive in unpredictable environments. As our Robotics Autonomy Engineer – Manipulation, you’ll lead the development and deployment of whole-body control and manipulation capabilities for humanoids and arm-equipped quadrupeds, turning real field scenarios into behaviors our robots perform reliably every day. You’ll be part of a deeply technical team advancing real-world robotic capabilities through cutting-edge research, simulation tools, and field validation. If building robots that can use their hands in the real world excites you, and you want to work where your code hits the ground (literally)—this is your role. This is Field AI. What You’ll Get To Do Develop whole-body control and manipulation capabilities for humanoids and quadrupeds with arms, coordinating base, torso, and arm motion as a single system Take specific field scenarios from concept to repeatable execution on hardware, interacting with the built environment and handling objects and tools Build learning based and model based manipulation policies and integrate them tightly with perception and locomotion Train manipulation policies from demonstration data and adapt vision language action (VLA) foundation models to our robots and tasks, then make them work on hardware Design grasp, contact, and force interaction strategies that hold up across varied objects, surfaces, and environments, with interaction forces kept safe for people and equipment nearby Harden behaviors so they succeed across repeated runs, changing lighting, object variation, and real site conditions Build simulation environments and evaluation pipelines for manipulation, and use field data to close the gap between simulation and the real world Deploy on physical robots and iterate rapidly through lab testing and field validation Work closely with locomotion, perception, and systems engineers to deliver end to end capabilities What You Have Master’s degree or higher in Robotics, Computer Science, Engineering, or related field (PhD strongly preferred) Deep expertise in robot manipulation or whole-body control 2+ years of experience developing and deploying manipulation behaviors on real robotic systems (preferred) Hands-on experience with robot arms, hands or grippers, and humanoid or mobile manipulation platforms Solid understanding of kinematics, dynamics, contact modeling, and force control Experience with manipulation planning and control, such as trajectory optimization or model predictive control (MPC) Experience with learning based manipulation, such as imitation learning, reinforcement learning (RL), or vision language action models Proficiency with simulation tools such as Isaac Lab, MuJoCo, or Drake Experience integrating perception for manipulation Strong Python and/or C++ development skills in Linux-based development environments Familiarity with machine learning frameworks (PyTorch, JAX, TensorFlow) A passion for building things that move in the real world The Extras That Set You Apart 3+ years of experience in an industry or startup robotics setting Experience with loco-manipulation on legged platforms or whole-body model predictive control for humanoids Experience fine tuning or deploying vision language action models on real robots Experience building teleoperation and demonstration data collection pipelines at scale Experience with cross embodiment or generalist manipulation policies that transfer across robots and tasks Experience with dexterous multi fingered hands, or with tactile and force torque sensing in closed loop manipulation Experience optimizing and deploying learning based controllers on resource constrained robotic platforms (ONNX Runtime, NVIDIA TensorRT, real time onboard inference) Publications or open-source contributions in manipulation, whole-body control, or robot learning (e.g., CoRL, RSS, ICRA, IROS) Familiarity with ROS/ROS2 or custom middleware for real-time control Experience debugging sim-to-real issues at scale