Robotics Engineering Intern: Tactile Manipulation

Robotics Engineering Internship:
Tactile Manipulation

Location:
Mountain View, CA

Compensation:
$4,000 - $5,000, depending on experience

Commitment:
Onsite, with meaningful hands-on lab time

We are an early-stage robotics company working at the intersection of robotic manipulation, tactile sensing, and physical AI.

We are building systems that help robots collect, interpret, and learn from contact-rich interaction data. This role is ideal for someone who wants to work close to real robots while thinking deeply about manipulation learning, tactile representations, robot data infrastructure, and modern ML pipelines.

We are looking for a motivated robotics engineering intern with strong Python skills and experience, coursework, or serious project work in robotics and machine learning.

You will help run manipulation experiments, collect high-quality robot and tactile data, and build tools that connect physical sensing systems to learning pipelines.

You will help with:

• Running manipulation experiments on a hands-on robotics test rig
• Collecting synchronized robot, sensor, camera, and tactile data
• Building Python and C++ tools for robot control, data logging, preprocessing, and automation
• Working with tactile and multimodal data representations for downstream ML models
• Supporting experiments involving diffusion policies, encoders, representation learning, or multimodal model workflows
• Structuring datasets for training, benchmarking, validation, and reproducibility
• Debugging issues across robot behavior, data quality, synchronization, ROS, sensors, cameras, and model inputs

You may be a strong fit if you have experience with several of the following:

• BS or MS student/recent graduate in engineering, robotics, computer science, or a related field
• Strong Python skills
• ROS or ROS 2 experience
• Experience with robotics data collection, ML experiments, or physical test rigs
• Exposure to robot learning, foundation models, encoders, representation learning, or multimodal models
• Comfort debugging real robot systems and messy physical data
• Strong engineering instincts and ability to work independently with guidance
• Long-term interest in robotics, physical AI, tactile sensing, or embodied intelligence

You will get real ownership over early robotics learning infrastructure at a startup. Your work will directly shape how we collect, structure, and use tactile data for manipulation and physical AI.

This is an early-stage, hands-on robotics prototype environment. The role requires meaningful onsite time in Mountain View for robot experiments, data collection, debugging, and validation. Candidates should be comfortable working in person with physical robots while also thinking about how the data flows into modern learning systems.