Computer Vision Engineer
On-siteLondon, England, United Kingdom
Job Summary
Design, train, and evaluate deep learning architectures alongside classical computer vision pipelines for feature tracking, optical flow, and spatial filtering. Develop robust mathematical pipelines for camera calibration and coordinate mapping to ensure accurate spatial outputs. Architect and containerise Python-based cloud microservices using Docker for production deployment, while assisting with cross-platform native binaries for desktop applications. Build intelligent, automated data-ingestion pipelines that utilise model-assisted pre-labeling to clean and version high-throughput training datasets. Define clean API boundaries and interface contracts to integrate core data science modules into downstream vertical applications. Collaborate with senior data scientists and vertical teams to translate product requirements into practical computer vision solutions from raw video ingestion to production inference.
Required Qualifications
- Foundational knowledge of classical computer vision (multi-view geometry, object tracking, spatial transformation)
- Foundational knowledge of modern deep learning architectures (object detection, semantic/instance segmentation, transformer-based vision models)
- Experience sourcing, structuring, training, and benchmarking deep neural networks using PyTorch or TensorFlow
- High proficiency in Python for prototyping, training, scripting, and deployment pipelines
- Practical, supporting capability to read, build, and debug existing C++ codebases
- Exposure to build management tools like CMake
- Familiarity with optimising model runtimes and inference execution graphs for real-time applications using TensorRT or ONNX Runtime
- Practical experience with Docker containerisation
- Experience with version control (Git)
- Practical experience with cloud platform execution (AWS)
- Must be based in our London office
Desired Qualifications
- Experience modifying, adapting, or designing custom neural network components (e.g., specialised backbones, attention mechanisms, or custom loss functions)
- A strong intuitive grasp or academic background in applied linear algebra and matrix calculus, particularly as it relates to 3D spatial transformations and projective geometry
- Experience or familiarity with native application development tools (Visual Studio, Qt Creator)
- Experience experimenting with or competing in open-source sports analytics datasets and benchmarks (e.g., SoccerNet, SportsMOT)
- A genuine interest in sports analytics, tracking technology, or elite human performance
Hiring someone like this?
Get your role in front of qualified candidates on Sorce.