Robotics Software Engineer Intern
Relling
Robotics Software Engineer Intern
Relling
Multi-modal AI for video-based perception
Hours: 40-60/week. Duration: A minimum commitment of 3–4 months is required.
Relling Systems (YC S25) builds ready-to-deploy AI robot workcells for US manufacturers. We design and test in San Francisco and put systems to work on factory floors. This role is in person in Potrero Hill, San Francisco.
Build a scoped robotics software feature or diagnostic tool with an engineer.
Write tests, inspect sensor data and reproduce runtime failures.
Document the interface, test result and remaining limits of your work.
Working toward a relevant degree or equivalent practical training.
Python or C++ fundamentals and a working project you can show.
Interest in software that has to work on real hardware; robotics experience is helpful but not required.
Apply with your resume and a short description of something you built. Project links are optional. Tell us what you owned, what failed and how you tested the fix. Coursework, student teams and personal projects count.
Relling develops AI models trained on raw video and multiple sensory signals (video, depth, IMU, audio, force, gaze) to understand and interact with the real world. The models learn from high-fidelity measurements rather than just text or images, enabling them to measure events directly and cope with challenges like blur and occlusion, especially in 3D scenes. Unlike typical text/image-based AI, Relling uses a broader sensor suite to train systems that perceive 3D structure and physical events, improving reliability in dynamic environments and defense-focused applications. The goal is to defend critical infrastructure by delivering AI that can reliably perceive, predict, and act in the real world, bridging digital intelligence with physical reality.
Explore more opportunities at Relling