Senior Mechanical Engineer – Robotic Manipulation & Learning

Amazon Inc.; San Francisco (CA); Hybrid; USD 140,000 - 190,000; Full time

Job summary

Amazon's Frontier AI & Robotics team in San Francisco is seeking a Member of Technical Staff - Mechanical Engineer to design and develop robotic manipulation hardware end-to-end, including grippers, EOAT, and data-collection fixtures, enabling AI researchers to collect high-quality demonstration data.

You will prototype quickly, iterate on end-effectors, integrate sensors, and work with AI researchers to understand which mechanical properties matter for learned manipulation policies.

Qualifications

• Bachelor's degree in mechanical engineering, mechatronics engineering, or equivalent.
• 3-5+ years designing mechanical systems for robotics, grippers, end-effectors, or complex electromechanical products.
• Proficiency in 3D CAD (SolidWorks, OnShape, or equivalent).
• Hands-on experience building prototypes, integrating hardware, and troubleshooting mechanical issues.
• Experience designing compact mechanisms with integrated actuation and sensing.
• Experience with rapid prototyping methods (3D printing, CNC, laser cutting, sheet metal).
• Familiarity with actuator/motor selection for compact, force-controlled applications.
• Experience with structural analysis and material selection for lightweight, functional parts.

Responsibilities

• Design grippers, end-of-arm tools, and low DoF multi-finger hands for robotic manipulation research and data collection.
• Develop UMI-style grippers and teleoperation hardware for collecting manipulation demonstration data.
• Design fixtures, jigs, and mounting systems for cameras, sensors, and manipulation test setups.
• Integrate tactile sensors, force/torque sensors, and cameras into compact gripper assemblies.
• Rapidly prototype and iterate on manipulation hardware — from concept sketches to functional grippers in days/weeks.
• Partner with AI researchers to understand what mechanical properties (compliance, sensing, geometry) matter for learned manipulation policies.
• Design data collection devices, such as UMI grippers, that enable repeatable, high-quality demonstration capture.
• Conduct mechanical testing of gripper performance (grasp force, compliance, durability, repeatability).
• Apply DFM/DFA to scale successful gripper designs from one-offs to small batches (10s-100s of units).
• Own cable routing, actuation, and sensing integration for compact end-effector designs.
• Support hands-on builds, debug sessions, and real-world manipulation experiments.
• Collaborate with controls, perception, and AI teams to ensure hardware meets research needs.

Skills

Hands-on prototyping

3D CAD

Prototype building

Hardware integration

Troubleshooting

Education

Bachelor's degree

Master's degree

Tools

SolidWorks

OnShape

Job description

Amazon's Frontier AI & Robotics team in San Francisco is seeking a Member of Technical Staff - Mechanical Engineer to design and develop robotic manipulation hardware end-to-end, including grippers, EOAT, and data-collection fixtures, enabling AI researchers to collect high-quality demonstration data.

You will prototype quickly, iterate on end-effectors, integrate sensors, and work with AI researchers to understand which mechanical properties matter for learned manipulation policies.