Computer Vision Engineer

We are looking for an experienced Computer Vision Engineer to help us make transport safer and greener, who thrives on solving complex problems and collaborating to drive product impact.

Salary: £55,000-£70,000k (if the advertised salary range is below your current expectations, we would still encourage you to apply. We are open to discussing the role and overall package in line with experience and scope)

Reporting to: Adam Fry (Engineering Manager, Sensor Hardware & Electronics)

Location: Primarily based in our London Office, with flexible and hybrid working (Wednesdays required with 2 days per week strongly recommended).

We are looking for a Computer Vision Engineer with experience in NVIDIA’s Edge AI stack to work on our core computer vision pipelines, powering real‑time inference used by authorities globally to make informed improvements to traffic systems.

In this role, you will develop and maintain the systems that power our AI traffic sensors - improving performance, accuracy, and efficiency across our GStreamer and Deep Stream-based pipelines. You will work closely with both researchers and hardware engineers to ensure that models are deployed effectively and run reliably in production.

This is a highly impactful engineering role focused on deep technical expertise, where you will own key parts of our vision stack and play a central role in improving performance, scalability, and long‑term evolution of our edge AI systems. Your work will substantially improve outcomes for customers, and ultimately road users around the world.

Your time will be spent roughly as follows:

• 75% - Core Engineering (pushing the boundaries of our Deep Stream and GStreamer pipelines, building new features, and deploying new deep learning models)
• 15% - Reactive debugging and support
• 10% Cross‑team initiatives (e.g. collaborating with cloud engineers on self‑learning algorithms, or building dashboards for surfacing new datasets you’ve developed)

This is a unique opportunity to work at the intersection of AI, hardware, and real‑world deployment - improving how thousands of sensors understand and interpret the world, and directly contributing to safer and more sustainable transport systems.

You are a hands‑on engineer with experience working on embedded computer vision systems and have a strong interest in how deep learning models perform in real‑world environments.

You are comfortable working within existing systems and improving them over time - whether that’s optimizing performance, simplifying complexity, or making systems more robust. You bring solid experience with NVIDIA’s vision stack and are confident working with GStreamer‑based pipelines.

You are also open and collaborative, and excited to share your knowledge with others - including engineers from different disciplines or with different levels of experience.

• Experience working with complex and custom NVIDIA Deep Stream and GStreamer‑based pipelines in production (e.g. nvargus, nvinfer, nvtracker)
• Proven ability to take ownership of complex vision systems, improving structure, maintainability, and enabling upgrades (e.g. Jet Pack / platform upgrades)
• Experience working under edge constraints (latency, compute, memory)
• Strong understanding of end‑to‑end video / vision systems, from camera input (incl. Optical performance, ISPs, and tuning) through to model inference and output

• Proficiency with Golang
• Experience deploying deep learning models into production, especially with custom layers or kernels
• Experience with production IoT systems
• Experience with MLOps practices
• Experience working in a start‑up or scale‑up environment
• Interest in sustainability, transport, or smart cities

You don't need to have done all of these things before, but to excel in this role, you will need to be keen to learn and comfortable working in a dynamic, fast‑paced environment. If you're close to what we're looking for, please consider applying. Experience comes in many forms, skills are transferable, and passion goes a long way.

• 30 minute screening interview.
• 1.5 hour system…