Robotics Perception Engineer Intern

About the role

The Robotics Perception Engineer Intern will develop and evaluate perception components on real robot data, build datasets and evaluation harnesses, run experiments, and report results. They will integrate their module into the live robot stack and demo it. Hands-on experience with pose estimation, VLMs, VLAs, 3D scene reconstruction, object detection, and multimodal learning is expected.

Responsibilities

• Develop and evaluate perception components on real robot data.
• Build datasets and evaluation harnesses; run experiments and report results.
• Integrate your module into the live robot stack and demo it.
• Test your work on real robots at the office.
• Opportunity to publish (for PhD interns).
• Attend Tier-1 industry conferences.

Requirements

• Pursuing MS/PhD in CS, EE, Robotics, or related field.
• Strong Python skills; experience with PyTorch and computer vision fundamentals.
• Evidence of applied work: projects, competitions, publications, or industry robotics experience.
• Comfortable taking ownership of tasks with light supervision.
• Excellent problem-solving skills.

Qualifications

• Must be legally authorized to do an internship in the United States for either 3 months or 6 months.
• Nice to have: Experience with sensor drivers, CUDA, 3D geometry, or multi-sensor fusion.