Full-Stack Robotics Engineer - Robotics & AI Systems
PathOn Robotics is a stealth-stage robotics startup based in the San Francisco Bay Area, building at the intersection of AI Agents + Robotics. Pre-seed funded.
Role Description:
This is a full-time onsite role for a Robotics Engineer in the SF Bay Area. You will work directly with the founder to build and ship robotics features end-to-end — from simulation to real robot deployment. This is a hands-on, high-ownership role. We need someone who can independently pick up tasks, research existing open-source solutions, integrate them into our system, and deliver without hand-holding. No visa sponsorship available. Must be a US citizen or green card holder. What You'll Work OnBuild and improve grasping, locomotion, and navigation pipelines in simulation (MuJoCo, ROS2/Gazebo, Isaac Gym) Integrate and deploy ML models (GraspNet, VLA, RL policies) for robot control Set up and maintain Docker-based environments for simulation, training, and inference Work with real robot hardware — calibration, control, sim2real transfer Contribute to fullstack platform infrastructure
Requirements:
- US citizen or green card holder (no visa sponsorship)
- Strong hands-on experience (5+ YOE) with ROS2, MuJoCo, or similar robotics frameworks
- Proficient in Python; comfortable with C++ is a plus
- Experience with at least one of: (1) Robot manipulation (grasping, motion planning, IK/FK), (2) RL for locomotion, (3) VLA / imitation learning
- Able to work with Docker, Linux, and GPU servers
- Self-directed: can independently research, prototype, and ship without constant guidance
- Bachelor’s or Master’s in Robotics, CS, EE, or related field
Nice to Have:
Experience with LeRobot, Isaac Gym, or MoveItPrior startup experience or open-source contributions Mechanical design or sensor integration (cameras, IMUs)
Why Join:
- Direct impact: shape the product from day one, working closely with the founder
- Real equity with transparency
- Accelerated learning across the full robotics stack — simulation, ML, hardware, platform
- No bureaucracy, no layers — just build and ship