Controls Engineer

About the role

ABOUT THE COMPANY

The datacenter buildout is the largest industrial project in human history. Gradient builds the autonomous robots that make it possible.

Partnered with the world's largest AI infrastructure companies and backed by the industry's best investors, we move fast and build full-stack systems that matter.

ABOUT THE ROLE

We're looking for a Controls Engineer to own motion from command to actuator. That means modeling and identifying the physical system, understanding exactly why an axis or arm is limited to the performance it has, and designing controllers that deliver sub-millimeter precision under real load. You'll be the person who can look at a Bode plot from the robot and explain what's in it. You'll also choose and source the motors, drives, and sensors your controllers run on, deciding where an off-the-shelf servo drive is the right answer and where we need something custom.

RESPONSIBILITIES

- Model and identify axes and arms using physical principles and measurement (e.g., swept-sine/chirp FRFs, step and noise-based ID, dynamic parameter identification), maintaining accuracy against hardware.

- Diagnose and resolve field issues including resonances, instability, non-linearities (friction, backlash), joint compliance, latency, sensor noise, and dynamic variability.

- Design and tune control loops: current, velocity, and position cascades, inverse-dynamics feedforward, notch and loop-shaping filters, and force/impedance control for contact tasks.

- Select and source motion and sensing hardware (motors, drives, encoders, force/torque sensors) and fieldbus modes. Commission commercial drives (Yaskawa, Kollmorgen, Elmo, Beckhoff, etc.) and specify custom drive/loop architectures where necessary.

- Develop test procedures to verify performance, and quickly iterate from failed tests to tuned revisions.

MINIMUM QUALIFICATIONS

- 3+ years' experience in controls engineering on physical electromechanical systems with control loops tuned on real hardware.

- Strong control theory fundamentals: frequency-domain analysis, stability margins, loop shaping, discrete-time effects, and sound judgment for controller selection (e.g., PID with feedforward).

- Demonstrated system identification skills: measured FRFs, model fitting, and controller design using those models.

- Experience with serial robot arms: forward and inverse kinematics (handling singularities, redundancy, joint limits), rigid-body dynamics, friction compensation

- Expertise in debugging nonlinearity and bandwidth-limiting problems and explaining their root causes.

- Fluency in Python for analysis and ability to write and review real-time control code in C, C++, or Rust.

- Ability to work on-site in San Francisco, 5 days/week (6 days/week during crunch time).

PREFERRED QUALIFICATIONS

- Experience with force/impedance control or other contact-rich robotic manipulation.