Perception Engineer: LiDAR & CV for Warehouse Autonomy

Dexory; Wallingford; Hybrid; GBP 60,000 - 90,000; Full time

Benefits offered by this job

Private health care

Life insurance

Income protection

Pension

Employee wellbeing support

25 days holiday + bank holidays

Meals onsite

Team events

Job summary

Dexory in Wallingford, Oxfordshire, is seeking a Perception Engineer with strong Python and C++ skills to develop perception pipelines for warehouse environments. You will work with LiDAR and camera data, implement both classical and learning-based approaches, and integrate with a ROS2 autonomy stack.

The role combines research and production with a hybrid work setup. Candidates should have solid fundamentals in 3D geometry, computer vision, and machine learning, and be comfortable traveling to

Qualifications

• Strong foundations in 3D geometry, sensor data, and perception.
• Experience with 3D point clouds, LiDAR, and cameras.
• Familiarity with ROS/ROS2 and perception pipelines.

Responsibilities

• Perception pipeline development for point cloud processing, object detection, and scene understanding in warehouse environments.
• Sensor data analysis: LiDAR scans and camera streams for debugging and improvement.
• Data collection and labeling to support datasets for classical and learned components.
• Collaborate with robotics engineers to integrate perception into the ROS2 autonomy stack.
• Run structured experiments and measure accuracy improvements across environments.
• Build and maintain tools for analysis, visualization, and debugging of perception.
• Stay current with research and contribute to the roadmap.

Skills

Python

C++

3D geometry

Computer vision

Machine learning

ROS2

Signal processing

Tools

ROS2

PCL

Open3D

OpenCV

PyTorch

TensorFlow

Job description

Dexory in Wallingford, Oxfordshire, is seeking a Perception Engineer with strong Python and C++ skills to develop perception pipelines for warehouse environments. You will work with LiDAR and camera data, implement both classical and learning-based approaches, and integrate with a ROS2 autonomy stack.

The role combines research and production with a hybrid work setup. Candidates should have solid fundamentals in 3D geometry, computer vision, and machine learning, and be comfortable traveling to