Location: Remote / Hybrid (UK-Based ONLY)
Type: Full-time
The Role
We're looking for a Senior Computer Vision Engineer to lead the development of real-time vision systems across multiple domains—military, security, logistics, and industry. You'll work end-to-end: from algorithm design and model training to deployment on edge hardware. If you're the kind of engineer who prefers building tools over writing papers, we want to hear from you.
What You'll Do
- Design, train, and optimise deep learning models for object detection, face recognition, tracking, and scene understanding
- Own the lifecycle of CV systems from data collection and annotation to deployment on embedded and edge devices
- Collaborate with hardware engineers to tune performance on platforms like NVIDIA Jetson, Intel Movidius, and Raspberry Pi
- Evaluate open-source models and datasets, and fine-tune for real-world robustness
- Integrate vision systems with AR displays, tactical UIs, or cloud dashboards depending on project needs
- Rapidly prototype and iterate using Python, OpenCV, PyTorch, TensorRT, and related tools
- Write clean, production-ready code with a focus on maintainability and scalability
Must-Have Skills
- 5+ years of experience in Computer Vision and Deep Learning
- Strong knowledge of CNNs, object detection architectures (YOLO, SSD, Faster R-CNN), and tracking methods
- Hands-on with Python, PyTorch, OpenCV, and at least one deployment framework (ONNX, TensorRT, OpenVINO, etc.)
- Experience with edge AI platforms (e.g., Jetson Nano/Xavier, Coral, RKNN, etc.)
- Ability to design and manage custom datasets, labeling pipelines, and performance benchmarking
- Solid engineering mindset—code, debug, test, repeat
- Excellent communication skills—able to explain complex things clearly, fast
Bonus Points For
- Experience in thermal/FLIR image processing
- AR/XR integration experience (Unity, WebXR, or custom HUD overlays)
- Background in real-time video streaming or low-latency pipelines
- Contributions to open-source CV projects or published GitHub repos
- Prior work in defence, industrial automation, or autonomous systems
Why Obsidian Reach?
We build AI that works in the real world—no hype, no hand-waving. Our clients span defense,
industry, and high-stakes operations where failure isn't an option. We move fast, work
smart, and stay close to the mission. If you're driven by impact, not ego, and prefer
building to posturing, you'll fit right in.