Senior Computer Vision/ Machine Learning Engineer

Senior Machine Learning Engineer - Computer Vision Location: UK We're looking for a Senior Machine Learning Engineer with strong Computer Vision and Psychical Face Recognition experience to take ownership of a complex capability from research through to real-world deployment. This is a genuinely hands-on role. You'll work across the full lifecycle of a production face recognition system - developing models, building evaluation frameworks and optimising deployment on edge GPU hardware. This isn't a research-only role. You'll be taking models from experimentation through to live, production systems and solving the real-world challenges that come with it. About the Role: You'll take ownership of the face recognition pipeline, including: * Psychical Face detection, alignment, quality filtering and embedding extraction. * Training and fine-tuning face recognition models. * Building robust 1:1 verification and 1:N identification evaluation frameworks. * Developing multi-camera Computer Vision pipelines. * Optimising models for edge deployment using ONNX and TensorRT , including FP16/INT8 quantisation. * Balancing accuracy, latency, throughput and GPU compute constraints. * Designing matching, gallery and enrolment infrastructure. * Benchmarking model performance and tracking improvements across versions. * Reviewing code and mentoring other engineers. What We're Looking For: * Face Recognition -Essential * You must have hands-on experience building and deploying production face recognition systems . * Computer Vision * Learned embeddings and open-set matching. * Modern face recognition approaches such as RetinaFace, SCRFD, ArcFace, CosFace or AdaFace . * Face detection, landmark alignment and embedding models. * 1:1 verification and 1:N identification . * Evaluating biometric systems using metrics such as TAR/FAR, FMR/FNMR, DET curves and Rank-N accuracy . * Evaluating model performance in challenging, real-world environments rather than relying solely on standard benchmarks. Technical Experience * Strong Python and PyTorch experience. * Production Computer Vision and Machine Learning experience. * ONNX and TensorRT deployment. * NVIDIA GPU inference and optimisation. * Understanding of batching, memory constraints, latency and throughput. * Linux, Docker, Git and CI. * C++ experience would be beneficial. Highly Desirable * NVIDIA DeepStream , GStreamer or Triton. * Face recognition in challenging environments, including low-resolution imagery, motion blur, occlusion and difficult lighting. * ANN/vector search, including FAISS or HNSW . * Multi-camera tracking or identity association. * Face quality assessment or template fusion. * Liveness or presentation attack detection. * Model optimisation, distillation or pruning. * Synthetic or augmented training data. The Person We're looking for someone with strong engineering judgement who can confidently answer a simple but important question: is this model ready for production? You'll be comfortable working independently, making technical decisions based on data and performance, and challenging assumptions when something isn't good enough. You'll also have the opportunity to help shape how complex Computer Vision systems are built, evaluated and deployed in real-world, performance-constrained environments. If you have genuine production Face Recognition experience and enjoy solving difficult Computer Vision problems beyond the research stage, this is an opportunity to have real technical ownership and impact.