Robotics Software Engineer, Manipulation

We are looking for a senior Robotics Software Engineer to join our clients Manipulation team. In this role you will own and drive development across the full manipulation stack; from grasp planning through trajectory optimization, motion control, and whole-body integration. You will work at the intersection of classical robotics and modern machine learning, translating algorithmic research into reliable, production-ready systems on our clients humanoid platform. We are looking for someone with strong fundamentals across the modular components of a manipulation pipeline, paired with the engineering discipline and software skill to build robust, production-ready implementations.

This role is focused on manipulation specifically- deep experience in locomotion, multi-robot/fleet coordination, or human-robot interaction and intent-signaling is not a match for what we're hiring for, even if the underlying robotics background is strong.

Your Role:

Develop and optimize the full manipulation pipeline, integrating perception inputs through grasping, trajectory optimization, motion planning, and control

Implement robust high-level control strategies for precise manipulation including force/motion control and visual servoin

Integrate the manipulation pipeline with the humanoid robot's whole-body controller and contribute to loco-manipulation development

Develop digital twins of manipulation scenes for algorithm development using simulators such as MuJoCo and Isaac Sim

Conduct rigorous testing in both simulated and real-world environment

Work closely with the machine learning team to help train and deploy models for advanced manipulation task

Stay informed about the latest research, leveraging both traditional methods and deep learning-based approaches

We're Looking For:

BS, MS, or PhD in Robotics, Computer Science, or a related field

5+ years of experience in robotics with a strong emphasis on high-level control and manipulation of robotic or humanoid arms (this role sits above actuator-level control, which is owned by a separate team.)

Strong background in motion planning, control theory, and optimization.

Demonstrated expertise implementing manipulation algorithms for tasks such as pick-and-place, bin picking, door opening, and similar behaviors deployed and validated on real hardware, ideally at production scale.

Experience with robotic end-effectors or multi-fingered hands in real-world deployment.

Proficiency in C++, Python, and relevant robotics libraries (MoveIt!, cuMotion, Drake, OpenRAVE)

Experience with trajectory optimization libraries (IPOPT, SNOPT, qpOASES).

Experience with computer vision algorithms, sensors, point clouds, segmentation, and object detection

Familiarity with ROS, LCM, or other middleware for robotic systems.

Bonus Qualification

Experience developing algorithms for contact modeling and force/torque estimation

Experience with machine learning for manipulation, including behavior cloning and reinforcement learning.

Publications in top-tier robotics conferences focusing on manipulation systems.

Experience in a humanoid robot startup environment.

#J-18808-Ljbffr