Location: Remote - (Must be EU based)
Salary: Base salary equity
We are hiring a Computer Vision Engineer to join a robotics company building intelligent, flexible robots for real manufacturing environments. This role is focused on production systems. You will be working on perception pipelines that run 24/7 on deployed robotic cells.
The robots follow a modular architecture where new capabilities are continuously added, such as object recognition, grasp point estimation, anomaly detection, and task specific behaviours. As the system scales, the ML and MLOps foundations become critical. This role exists to own and extend those foundations.
What you will be doing
- Design, build, and deploy 2D and 3D computer vision systems used in live production environments. This includes image classification, object detection, semantic and instance segmentation, metric learning, and smart filtering.
- Take models end to end, from data and training through to deployment, optimisation, and monitoring.
- Contribute directly to an in house MLOps platform that supports data ingestion, experiment tracking, model versioning, deployment, and observability across multiple robotic capabilities.
- Work closely with robotics and hardware focused teams and help ensure models run efficiently and reliably on edge and production hardware. Over time, this includes model conversion and optimisation using tools such as ONNX and TensorRT.
What we are looking for
This is a production engineering role. We are looking for someone who has built ML systems before.
Non-negotiable experience:
- Hands-on experience working with visual data in production systems (2D and/or 3D computer vision).
- Proven production ML experience: you have taken models from training through deployment and supported them in live environments.
- Strong Linux fundamentals, including working over SSH and operating production infrastructure.
- You have built MLOps systems, not just used them. This includes ownership of data pipelines, experiment tracking, model versioning, deployment, and monitoring.
- Solid understanding of how models actually work under the hood. You are comfortable reasoning about backpropagation, gradients, network architectures, and debugging model behaviour when things go wrong.
- Experience with robotics, autonomous systems, or other edge-deployed ML.
- Synthetic data generation and the ability to design efficient data collection strategies.
- Model conversion and optimisation workflows using ONNX and/or TensorRT.
- Experience with ROS, Kubernetes, and cloud platforms.
This role offers a rare combination of real-world impact and deep technical ownership. You are not optimising isolated models or working on disconnected experiments. You are helping define the perception and MLOps foundations for intelligent robotic systems that are already operating in production and will continue to scale over time.
Engineers work in pods with clear ownership and the opportunity to grow into leading entire problem areas. Progression and compensation are tied to impact rather than tenure. The environment is fast-moving and flexible, with intense periods of work when needed, and a strong emphasis on transparency and alignment.
This is a role for someone who wants to see their work move quickly from code to real machines on real factory floors.
