Computer Vision Lead

Lumyn; Greater London; Hybrid; GBP 84,000 - 125,000; Full time

Benefits offered by this job

Stock options

Job summary

Lumyn, in partnership with a defence technology company, seeks a hands-on Computer Vision Lead for autonomous drone platforms. You will define the roadmap, build the team, and also write and ship code that runs in real time on edge hardware during outdoor flights.

You will own the CV roadmap, develop object detection and tracking, SLAM, and ensure models run on embedded compute. The role offers hybrid work from London or Munich, with stock options.

Qualifications

• Degree in CS, robotics or related field; PhD beneficial.

Responsibilities

• Own the computer vision roadmap and tie milestones to product releases.
• Build real-time detection and classification of objects seen from the air.
• Track multiple objects across frames and cameras.
• Develop visual SLAM and visual navigation for outdoor flight.
• Get models running on embedded compute fast enough for flight.
• Benchmark every change on real flight data and in simulation.
• Hire computer vision engineers and grow them into a strong team.
• Set the standard for code review and documentation.

Skills

Applied computer vision

Deep learning (PyTorch/TensorFlow)

Object detection (YOLO/Faster R-CNN)

Multi-object tracking

Visual SLAM

C++ and Python

Team leadership

OpenCV

CUDA

ROS (Robot Operating System)

Education

Bachelor's degree in CS/Robotics or related field

PhD (preferred)

Tools

ROS

Job description

Lumyn has partnered with a venture-backed defence technology company that builds autonomous drone systems. They are looking to bring on a Computer vision lead.

The role

You'll lead computer vision across the company's drone platforms. It's a hands-on lead role. You set the roadmap and build the team, and you still write and ship code. Your code runs in real time on edge hardware, on a moving platform.

What you'll do

• Own the computer vision roadmap and tie its milestones to product releases.
• Build real-time detection and classification of objects seen from the air.
• Track multiple objects across frames and cameras.
• Develop visual SLAM and visual navigation for outdoor flight.
• Get models running on embedded compute fast enough for flight.
• Benchmark every change on real flight data and in simulation.
• Hire computer vision engineers and grow them into a strong team.
• Set the standard for code review and documentation.

What you'll bring

• A degree in computer science, robotics or a related field. A PhD is a plus.
• Computer vision you have shipped into a real product.
• Deep learning with PyTorch or TensorFlow.
• Object detection models such as YOLO or Faster R-CNN.
• Multi-object tracking.
• Visual SLAM and structure from motion.
• Strong C++ and Python. ROS is a plus.
• OpenCV and CUDA, with real-time video processing.
• Experience leading engineers, as a team lead or as the technical lead on a vision product.

What's on offer

• €100,000 to €150,000 base, plus stock options.
• Hybrid working from Munich, or London
• Ownership of the vision team and its roadmap.