Robotics AI Intern: VLA Data & Vision Models

Rival; Zürich; Hybrid; CHF 20.000 - 33.000; Teilzeit

Zusammenfassung

RIVR, part of Amazon, is looking for an AI Intern to help build data pipelines and train Vision-Language-Action models for robotic systems in Zürich. You will work at the intersection of data engineering and model training, solving the data bottleneck in embodied AI, and collaborating with senior engineers to curate and visualize multi-modal data generated by our fleet.

You will gain hands-on experience with state-of-the-art VLA models, develop tools to visualize data and debug performance, and

Qualifikationen

• Proficiency in Python and deep learning frameworks, preferably PyTorch.
• Hands-on experience with deep learning for computer vision.
• Familiarity with data manipulation libraries such as NumPy, Pandas, and OpenCV.
• Experience working with multi-modal data and datasets is a plus.
• Eagerness to learn and contribute in a collaborative team.

Aufgaben

• Support processing, curating, and analyzing multi-modal sensor data for training VLA models.
• Gain hands-on experience with state-of-the-art VLA models.
• Develop software tools to visualize data and debug model performance.
• Collaborate with senior engineers to implement data strategies for robustness of robotic systems.
• Assist in integrating software components to evaluate algorithms in simulation and on hardware.

Kenntnisse

Python

PyTorch

NumPy

Pandas

OpenCV

VLA models

Ausbildung

BSc in Computer Science / Robotics / ML

Jobbeschreibung

RIVR, part of Amazon, is looking for an AI Intern to help build data pipelines and train Vision-Language-Action models for robotic systems in Zürich. You will work at the intersection of data engineering and model training, solving the data bottleneck in embodied AI, and collaborating with senior engineers to curate and visualize multi-modal data generated by our fleet.

You will gain hands-on experience with state-of-the-art VLA models, develop tools to visualize data and debug performance, and