Optical Sensing Intern - Robot Perception

At Rhoda AI, we’re building the next generation of generalist intelligent robots. We own the full robotics stack from high-performance hardware and robot systems to the infrastructure and state-of-the-art foundation world models that control our robots. Our robots are designed to be generalists capable of operating in complex, real-world environments and handling long-tail edge cases, made possible by our cutting edge research and end-to-end system design. We've raised over $450M and are investing aggressively in model research, infrastructure, hardware development, and manufacturing scale-up to make generalist robotics a reality.

Level: Internship or co-op, MS students; 4–6 months duration - starting as soon as possible

You’ll help decide how our robots see. Our robots work around people in real warehouses, and every sensor has conditions that defeat it: bright sunlight, shiny floors, glass, dust, heat. Your experiments will tell us which settings to ship, where each sensor breaks and why, and what we should do about it. You’ll work directly with our perception and hardware engineers, and your findings go straight into sensor choices, settings and the requirements our detection algorithms are built against. This is a hands-on internship for someone who understands optical physics and wants to see their work running on real robots.

WHAT YOU’LL DO

- Design experiments that answer real engineering questions, such as which sensor settings work best for each job, and whether a sensor still sees a person in direct sunlight

- Explain why a sensor fails, from the physics up: ambient light, dynamic range, multipath, reflectance, angle of incidence, heat

- Propose and test what to do about each failure: a different setting, a filter, a mounting change, or a check the software can run

- Build the scenes, targets and scenarios that reproduce real-world conditions in the lab, and in the field when the lab isn’t enough

- Use sensor SDKs and firmware to sweep and tune settings such as exposure, gain, frame rate, range modes and filtering

- Write Python to process images and point clouds and turn raw data into clear numbers: range, accuracy, noise, dropout

- Present findings and recommendations to engineers who will act on them

WHAT WE’RE LOOKING FOR

- Pursuing an MS in Optics, Electrical Engineering, Physics, Computer Vision or Robotics, with research or lab experience designing experiments

- Strong grasp of optical sensing physics: image formation, noise, dynamic range, time-of-flight and LiDAR principles

- Hands-on with cameras, LiDAR, time-of-flight or other optical sensors in a lab or on a robot

- Comfortable with sensor SDKs, drivers and configuration

- Python for numerical analysis and image or point-cloud processing (NumPy, OpenCV, Open3D or similar)

- Curiosity to dig past “it doesn’t work” to the cause, and creativity in finding fixes

- Explains results clearly, in writing and with plots, to people outside your specialty

NICE TO HAVE

- Camera or sensor calibration

- Thermal or infrared imaging

- Optical modeling or sensor simulation

- ROS 2 and recorded-data tools