Senior Computer Vision Engineer — eKYC & ML Systems

About Blnkblnk is a fintech company on a mission to enable financial inclusion through point-of-sale financing. Less than 4% of Egyptians have access to credit cards. Many consumers rely on cash savings or are forced to borrow from informal lenders at high interest rates to purchase products and services. We’re changing this by enabling consumers to access credit within minutes at their preferred merchants.About The Roleblnk is looking for a Senior Computer Vision Engineer to own and improve its eKYC and identity verification systems. The role combines hands-on expertise in computer vision and deep learning with production engineering, inference optimization, and technical leadership.The successful candidate will own the eKYC technical roadmap, maintain and enhance existing services, and lead a mid-level Computer Vision Engineer in developing, evaluating, and deploying reliable ML solutions.Key Responsibilities eKYC & Identity Verification: Own and improve face verification, face alignment, liveness detection, OCR, document processing, and identity fraud detection systems. Model Development: Develop, fine-tune, evaluate, and optimize computer vision models to improve accuracy, robustness, latency, and operational reliability. Training Automation: Build reusable model-training and evaluation pipelines covering dataset preparation, experiment tracking, benchmarking, model selection, and artifact versioning. High-Performance Deployment: Develop and optimize production inference systems using C++ (C++17 or later), TensorRT, ONNX Runtime, CUDA, and OpenCV. GPU Optimization: Profile inference workloads, optimize GPU memory usage and execution, and implement or optimize custom CUDA kernels where necessary. Production Ownership: Maintain the eKYC server, troubleshoot production issues, improve observability, and implement regression testing, monitoring, and safe model releases. Technical Leadership: Lead and mentor the mid-level Computer Vision Engineer, conduct code reviews, define engineering standards, and coordinate technical priorities with the broader ML team. Required QualificationsStrong experience in computer vision and deep learning, including model training, evaluation, and optimization.Hands-on experience with face recognition or verification, OCR, document processing, or fraud detection.Strong Python and PyTorch skills, with experience deploying models into production.Strong C++ development skills and experience building or optimizing production inference systems.Practical experience with TensorRT, ONNX Runtime, CUDA, or comparable inference optimization technologies.Experience profiling and optimizing inference latency, throughput, and memory usage.Experience building reproducible training, evaluation, and deployment pipelines.Ability to mentor engineers, review technical implementations, and own production systems independently.Preferred QualificationsExperience with face alignment, face embeddings, biometric verification, and active and passive liveness detection.Experience implementing custom CUDA kernels and optimizing GPU inference.Familiarity with model quantization, batching, mixed precision, and other inference optimization techniques.Experience with Docker, FastAPI, CI/CD, model monitoring, and production observability.Experience building automated pipelines for training, benchmarking, and deploying multiple model architectures.Success CriteriaReliable ownership and continuous improvement of the eKYC platform.Measurable improvements in verification accuracy, fraud detection, inference performance, and service reliability.Reproducible training and evaluation pipelines that accelerate model development.Effective mentorship and increased technical independence of the mid-level engineer.Maintainable, optimized, and well-documented production ML systems.