Senior Computer Vision Engineer - Robotic Manipulation

The AI Teams at Agile Robots are looking for a Senior Computer Vision Engineer (m/f/d) Robotic Manipulation, who will build vision and software components that let our robots recognize and handle objects in close-to-industry applications. The stack combines in-house software with external research models, so you will help bring new results onto real robots without breaking what already runs.

Responsibilities

• Vision Components: Develop and integrate learning-based vision components, such as object detection, segmentation, and pose estimation, that give robots the input they need to grasp and manipulate objects.
• Software Architecture: Design and implement software components, architectures, and applications for AI-enhanced robotic systems.
• Robot Integration: Bring applications onto a changing range of robots and sensors, including dual-arm and humanoid platforms.
• Research Transfer: Follow current research in computer vision and robot learning, and evaluate which external models and components are ready to integrate into the stack.
• Stack Maintenance: Review, test, and maintain the shared software stack so that existing applications keep running as new components are added.

Essential Skills

• Background: Degree in robotics, computer vision, computer science, electrical or mechanical engineering, or a related field, or equivalent experience, with around four years of professional experience building robotics or vision software.
• Computer Vision: Hands-on experience training, adapting, and deploying current learning-based vision models, including pretrained vision and vision-language models, on real camera data.
• ROS 2: Professional experience developing robotics software with ROS 2.
• Programming: Production experience in C++ and Python, working with CMake, Git, and shell scripting.

Beneficial Skills

• Vision-Based Manipulation: Experience using camera input to guide grasping or manipulation, for example grasp pose estimation or visual servoing.
• Humanoid Platforms: Experience developing software for dual-arm or humanoid robots.
• Motion Control: Working knowledge of robot motion planning and control.
• Task Logic: Experience structuring robot task logic with state machines or behavior trees.