> Markdown version of [/jobs/ext/4772832-robotics-software-engineer-motion-planning](https://www.wearedevelopers.com/jobs/ext/4772832-robotics-software-engineer-motion-planning). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md). --- # Robotics Software Engineer, Motion Planning - **Company:** Path Robotics - **Location:** United States - **Experience:** Experienced - **Contract:** Permanent contract - **Skills:** Training Data, Unit Testing, Continuous Integration, Linux, Python (Programming Language), Kinematics, Machine Learning, Reinforcement Learning, Multithreading, Robot Operating System, Git, Information Technology, Data Analytics - **Published:** November 18, 2026 - **Apply:** https://startup.jobs/robotics-software-engineer-motion-planning-path-robotics-8730381 ## About the Role * 2+ years building robotics software in modern C++. Python is a strong plus. * Strong foundations in linear algebra, 3D geometry, kinematics, and optimization. * You've built software that runs on physical robot hardware. * Fluent developing on Linux with Git, with solid testing habits (unit tests, CI/CD). * You explain planning tradeoffs clearly in writing and design reviews, and work closely with perception, controls, and field teams. * A Bachelor's in Computer Science, Robotics, Electrical or Mechanical Engineering, Mathematics, or equivalent experience. Preferred * Experience with both sampling-based and optimization-based planning, and the judgment to know when each fits. * Experience applying machine learning to planning or control, such as learned proposals or costs, imitation or reinforcement learning, or sim-to-real. * Planning or collision checking under uncertainty, or online trajectory correction and replanning. * Multithreading, concurrency, and profiling of performance-critical code. ## Description * Design, build, and improve motion and trajectory planning for robot manipulators, from kinematics and inverse kinematics to collision checking and trajectory optimization, producing fast, feasible motion that meets several objectives at once. * Model collisions under uncertainty from imperfect sensing, calibration, and scene models, using both geometric and learned models, balancing safety margins with motion efficiency. * Modify trajectories to correct, adapt, and replan motion as sensor feedback and task requirements change, while keeping it collision-free and feasible. * Plan for systems with multiple degrees of freedom, including multiple robots working the same part, and sequence their tasks. * Improve classical planners with learning-based and data-driven methods, such as learned initial guesses for optimization, learned costs, or offline optimization that speeds up online planning. * Take learned models from prototype to production, with the validation, fallbacks, and monitoring that keep motion safe and correct. * Turn data from our production fleet into training data, benchmarks, and planning metrics, validate gains in simulation and on live robotic cells, and diagnose failures to their root cause., * Daily free lunch to keep you fueled and connected with the team