Robotics Manipulation R&D Engineer

Role Overview

The Robotics Manipulation R&D Engineer will bridge the gap between cutting-edge academic research and real-world manufacturing solutions. The primary focus will be tackling the manipulation of deformable objects.

What You’ll Do

• Develop and Deploy: Design, train, and deploy novel robotic manipulation policies (e.g., RL, imitation learning, diffusion policies) specifically tailored for handling fabrics and highly deformable textiles.

• End-to-End Pipelines: Architect and implement end-to-end manipulation pipelines that seamlessly integrate perception, state estimation, planning, and control for fabric manipulation.

• Sensor Integration: Integrate multi-modal sensors (e.g., tactile sensors, advanced 3D vision, force-torque) to achieve robust state estimation and physics-aware reconstruction of deformable materials.

• Sim-to-Real Transfer: Build and leverage robotic simulation environments (including deformable physics simulations) at scale to train models and successfully transfer them to physical hardware.

• Production Engineering: Write, optimize, and maintain reliable, production-level Python and C++ code to deploy your models onto real-world robotic platforms.

• Cross-Functional Collaboration: Work closely with mechanical engineers, software teams, and product stakeholders to iteratively design custom end-effectors, tackle system bottlenecks, and translate operational needs into technical roadmaps.

Minimum Qualifications

• Master’s or PhD degree in Robotics, Computer Science, Mechanical/Electrical Engineering, or a related technical field.

• 3+ years of hands-on experience in robotic manipulation, control systems, or applied embodied AI development.

• Strong proficiency in Python and C++ within a production or advanced prototyping environment.

• Proven experience deploying machine learning, reinforcement learning, or deep learning models onto real-world physical robots.

• Proficiency with modern deep learning frameworks (PyTorch, JAX, TensorFlow) and robotics systems (ROS/ROS2, modern sim environments).

• Solid understanding of kinematics, dynamics, motion planning, and systems-level robotic architecture.

Preferred Qualifications

• The Big Bonus: Direct hands-on research or industry experience with deformable object manipulation (e.g., fabrics, textiles, cables, or soft matter).

• Ph.D. in AI, Robotics, Machine Learning, or a related field with a focus on robotic manipulation.

• Strong foundational understanding of statistics and linear algebra relevant to deep learning and robot state estimation (e.g., Kalman filters, Gaussian processes).

• Familiarity with physics-aware simulation tools tailored for deformables.

• Proven track record of applying robot learning techniques (Sim-to-Real, RL, Imitation Learning, Action-Conditioned World Models) to solve complex, tangible problems.

• Publishations at top-tier ML, computer vision, or robotics venues (e.g., CoRL, ICRA, RSS, NeurIPS, CVPR).

• A product-oriented mindset with the ability to navigate tradeoffs between pure research exploration and scalable, reliable production deployment.

Pay

N/A

Schedule

Primarily onsite in our R&D Innovation Center in Newark, CA, with some work-from-home flexibility.

Benefits

N/A

Qualifications

• Full authorization to work in the United States.

• This role is not eligible for new employment visa sponsorship.

• Candidates must have full authorization to work in the United States.

• The CreateMe office is in Newark, CA and this role is not eligible for relocation assistance.