Robotics Vision & Perception Engineer (m/f/d)

Batch Robotics is revolutionizing how robots operate in factories — not through rigid programming or teaching by human demonstration, but with smart perception and AI intelligence. We are a team of 4 motivated people with +25 years of experience in industry and robotics. Our mission is to make robots smarter and more flexible, enabling batch-size-1 automation for surface treatment economically. Our technology helps factories in Europe to remain competitive with intelligent robotics.

We are now looking for a Robotics Vision & Perception Engineer (m/f/d) with technical excellence and entrepreneurial drive to join us in developing the next generation of AI-driven robotic systems!

Multi-Camera & Sensor Fusion: Architect, integrate, optimize and evaluate multi-camera vision and other sensor hardware perception systems. This includes developing robust multi-sensor calibration, synchronization, and visual-inertial or multi-modal fusion strategies that withstand environment variability and industrial noise of factories.

Data Pipelines & ML Ops: Build end-to-end data pipelines spanning real-world vision and sensor (e.g. force, structured light, laser scanners) data ingestion, dataset management, and synthetic data generation via simulation tools.

3D Perception & Shape Reconstruction: Design, implement, and optimize computer vision and perception sensor software algorithms and ML systems to automate the generation of a digital twin of the workspace and workpiece under real-world factory constraints.

Factory Integration: Deploy trained models into low-latency, production-grade robotics systems, ensuring real-time inference and seamless communication of the perception stack with robot control and policy pathways.

Evaluation & Industrial Robustness: Run permanent, rigorous evaluation experiments, failure mode analyses, and benchmarking against accuracy and latency thresholds to guarantee fail-safe operation on the physical factory floor.

Research Background: Master’s degree in CS, Engineering, or a related field, alongside 3+ years of industrial experience or a completed Ph.D. in computer vision. You are capable of transferring new research papers in working code and can optimize for industrial-grade reliability.

Core Technical Stack: Deep, hands-on experience with 3D point cloud data and geometric libraries (e.g. Open3D), working with concepts like geometric registration, shape reconstruction, spatial queries, and surface segmentation.

Coding Excellence: Strong proficiency in Python with agentic programming; CUDA, C++ and/or Rust is an advantage.

Hands-On Attitude: Proven track record of deploying industrial vision systems and handling noisy, temporal, or large-scale data. Comfortable deploying systems on-site in the field yourself and continuously improving the hardware-software perception stack.

Soft Skills: Entrepreneurial drive, solution-oriented mindset, high agency and the ability to take direct responsibility in a fast-paced startup environment.