Robotics Algorithms Engineer, Manipulation & Control

Build the Path Forward

At Path Robotics, we're attacking a trillion dollar opportunity - doing things that have never been done before to support an industry hurting from a lack of skilled labor. Big, hard problems are what Path tackles every day, and our people are our greatest asset to get that job done. Our intelligent, hardworking team of people do the impossible every single day, yet remain incredibly kind, humble, and always ready to support one another.

We are building an elite robotics team to solve real-time, high-accuracy manipulation - making a robot arm and end-effector do precise physical work in the real world, from live sensor feedback, on platforms that are never perfectly calibrated. We are looking for an exceptional hands-on engineer to own the control-and-estimation heart of that system: the closed loop that turns what the robot senses into accurate motion, in real time, robustly, on real hardware - and to set its technical direction as it grows.

What you'll do

• Own the real-time closed loop from sensing to motion, hands-on: state estimation, feedback and predictive control, and the correction of end-effector trajectories to hit accuracy targets under uncertainty, noise and latency.
• Make hard, honest engineering calls about what can and cannot be estimated from the available sensing, and design systems that know their own uncertainty and fail safely.
• Build robust hard-real-time software that runs reliably on physical robots, not just in simulation.
• Set the technical direction toward learning-based and whole-body control.

Who You Are

• Deep, first-principles real-time control and state estimation - model-predictive control, Kalman/particle/moving-horizon estimation, observability analysis.
• Manipulation robotics - arm kinematics and dynamics, trajectory generation and correction, coordinate frames, hand-eye and extrinsic calibration.
• Sensor-guided motion / visual servoing - closed perception-to-control loops on real hardware (vision- or sensor-guided manipulation)
• Track record of real-time robotic systems, built and run on hardware.
• Depth in robot perception - geometric/model-based vision, 3D sensing, point-cloud registration, calibration - to define what your control loop needs from perception and to reason about how perception error propagates into motion.
• Modern, production C++ for real-time systems.
• ROS / ROS 2 and the discipline of hard-real-time software - deterministic loops, timing budgets, buffered concurrency, thread-safety, debugging on real hardware.