Robotic Software Engineer - (Senior, Staff, Lead) Manipulation

Ghost Robotics•Philadelphia, PA

19h•$140,000 - $185,000•Onsite

About The Position

We're a robotics company building autonomous systems that operate in complex, dynamic environments. We're hiring a Manipulation Engineer to design, implement, and deploy manipulation algorithms for arms mounted on dynamic legged robots operating in the real world. You'll build task and motion planning, grasping, and control pipelines in real-time software, and own their performance all the way down to hardware. This role suits engineers who thrive on high-velocity problem solving, deep technical ownership, and hands-on testing and validation. This is one role, open at the Senior, Staff, or Lead level. The responsibilities below reflect the core of the role at every level. The selected candidate will be placed at the level commensurate with the skills, scope, and experience they demonstrate through the interview process, and the offer will align with that level. Senior and Staff are individual-contributor positions. Lead carries the same technical scope plus people-leadership responsibilities, for candidates who want them.

Requirements

A strong background in robotic manipulation, with hands-on experience on multi-DOF arms in real systems.
Strong foundations in motion planning, control theory, optimization, and dynamical systems
Deep understanding of coordinate transformations, forward/inverse kinematics, Jacobians, and robot dynamics
Experience with motion planning tools (e.g., MoveIt, OMPL, or custom optimization-based planners)
Experience with multi-body dynamics, modeling, and simulation (e.g., MuJoCo, Gazebo, Isaac, Bullet/PyBullet)
Experience with ROS 2 and real-time middleware
Proficiency in modern C++ (C++17/20) and Python
Experience with Unix/Linux environments and software engineering best practices (version control, CI/CD)
Master’s or PhD in Robotics, Mechanical Engineering, Electrical Engineering, Aerospace Engineering, Computer Science, or a related field.
3+ years of hands-on experience in robotic manipulation, specifically with multi-DOF (Degree of Freedom) arms.
Strong foundations in motion planning, control theory and optimization, along with experience in dynamical systems.
Deep understanding of coordinate transformations, forward/inverse kinematics, Jacobians, and robot dynamics.
Experience with motion planning software tools (e.g., MoveIt, OMPL, or custom optimization-based planners).
Experience with multi-body dynamics, modeling, and simulation (e.g., MuJoCo, Gazebo, Isaac, Bullet/PyBullet).
Experience with ROS 2 and real-time middleware.
Proficiency in modern C++ (C++17/20) and Python for development and tooling.
Experience with Unix/Linux environments and software engineering best practices (version control, CI/CD).

Nice To Haves

Experience with legged or humanoid robots
Experience applying Reinforcement Learning, Vision Language Models, or Vision-Language-Action models to robotic decision-making and task planning
Familiarity with 3D perception, point cloud processing, and vision-based feedback in manipulation loops
Background in whole-body control frameworks (operational space control, MPC, and similar)
Experience with force-feedback control or hand-eye calibration
Publications or significant open-source contributions in robotics or machine learning
Demonstrated ability to lead technical efforts and mentor engineers

Responsibilities

Architect end-to-end manipulation frameworks, from high-level task planning down to perception and low-level joint control
Develop trajectory planning, obstacle avoidance, and real-time motion generation
Implement grasping strategies using proprioception and vision, for rigid and deformable objects
Use physics-based simulators to develop and validate manipulation policies with high-fidelity transfer to hardware
Write clean, maintainable C++ and Python, and debug system performance across simulation, hardware experiments, and fleet data
Collaborate with mechanical, perception, embedded, and systems teams to ensure end-to-end performance and robustness
Contribute to long-term architectural decisions
Owns a defined technical area of the manipulation stack and delivers independently end to end (Senior)
Mentors less-experienced engineers (Senior)
Track record of shipping manipulation software to real hardware (Senior)
Drives technical direction and architecture across the team's area (Staff)
Sets engineering standards and multiplies the output of others (Staff)
Demonstrated influence beyond their own deliverables, across teams or an entire manipulation stack (Staff)
Staff-level technical scope plus people-leadership responsibilities: direct reports, hiring, performance, and team planning (Lead)
Demonstrated ability to grow engineers and run a healthy, productive team (Lead)