Controls Engineer Intern (Spring-Summer 2026)

Deftai; San Francisco (CA); Hybrid; USD 48,000 - 69,000; Full time

Benefits offered by this job

Daily meals stipend

Job summary

Deftai in San Francisco, CA is seeking a Controls Engineer Intern to work on low-level control and motion systems for bimanual wheeled robots in production environments. You will ship code that runs on real hardware, debug edge cases on the factory floor, and directly impact robot performance.

The role covers real-time supervisory control, inverse kinematics improvements, sensor fusion for mobile base localization, and integration of new actuators with safety features.

Qualifications

• Strong fundamentals in classical control theory with hands-on hardware experience.
• Proficiency in Python and C/C++ for real-time embedded or robotic apps.
• Hands-on experience with inverse kinematics solvers and simulation environments.
• Experience with rigid-body kinematics and coordinate frame transformations.
• Experience with sensor fusion techniques using IMU, encoders, etc.

Responsibilities

• Design and implement real-time supervisory control for smooth mode transitions.
• Improve inverse kinematics solvers for multi-arm manipulation and robustness.
• Develop sensor fusion pipelines for precise localization and orientation control.
• Integrate new actuators and configure motor control modes with safety limits.

Skills

Python

C/C++

Control theory

State estimation

Kalman filtering

Tools

PyBullet

MuJoCo

Drake

ROS/ROS2

Job description

About This Role

Our robots operate in real production environments where control reliability isn't optional — it's the difference between a deployed system and a demo. As a Controls Engineer Intern, you'll work directly on the low-level control and motion systems that make our bimanual wheeled robots safe, precise, and production-ready. This is not a paper-reading internship. You'll ship code that runs on real hardware, debug edge cases that only surface on the factory floor, and see the direct impact of your work on deployed robot performance.

What You’ll Do

• Design and implement real-time supervisory control systems for smooth handoff between autonomous and teleoperated modes, including trajectory blending and forward-consistent motion during transitions.
• Improve inverse kinematics solvers for multi-arm manipulation — resolving singularities, boundary failures, and ensuring robust solutions across the full reachable workspace.
• Develop sensor fusion pipelines (IMU, odometry, vision) for precise mobile base localization and orientation control under tight angular tolerances.
• Integrate new actuators for bimanual arm systems — including motor driver communication, torque/position/velocity control mode configuration, gain tuning, and safety limit implementation

Required Qualifications

• Strong fundamentals in classical control theory (PID, impedance/admittance control, state estimation) with hands-on experience implementing controllers on physical hardware.
• Proficiency in Python and C/C++ for real-time embedded or robotic applications.
• Hands-on experience with inverse kinematics solvers and simulation environments (PyBullet, MuJoCo, Drake, or similar), including practical familiarity with IK libraries or custom Jacobian-based implementations.
• Experience with rigid-body kinematics, coordinate frame transformations (homogeneous transforms, SE(3), quaternion representations)
• Experience with sensor fusion techniques (Kalman filtering, complementary filters) using IMU, encoders, or similar proprioceptive sensors.

Preferred Qualifications

• Prior experience with ROS/ROS2, CAN bus communication, or motor control interfaces.
• Familiarity with trajectory smoothing and interpolation methods (spline-based planning, Ruckig, or similar time-optimal trajectory libraries).
• Experience with wheeled or mobile robot platforms, including human in the loop data collection framework
• Meaningful ownership from day one — your code ships to production robots.
• Direct mentorship from founding engineers working at the frontier of robot learning and deployment.
• Opportunity to co-author publications if research contributions warrant it.
• Daily meals stipend.

Expected Compensation

• $35 – $50 per hour

The pay offered for this position may vary based on several individual factors, including job-related knowledge, skills, and experience. The total compensation package may also include additional components/benefits depending on the specific role. This information will be shared if an employment offer is extended.

Internship Duration

• Opportunity to extend duration.
• Opportunity to convert to full-time employment.

Interview Process

• Intro call + 1 virtual technical interview

Onsite Work Requirement

This is a full-time, five-day-a-week onsite role in our San Francisco, CA, office. Please confirm if you can meet this onsite requirement.

Are you a US Citizen, Green Card Holder or Valid Visa Holder (H1-B, O-1, TN, E-3, OPT, etc)?