Computer Vision Engineer
About us:
Farsight Vision develops AI-powered geospatial intelligence solutions for defence and security operations. The platform transforms raw aerial and sensor data into real-time 3D environments, actionable analytics, and AI-assisted operational scenarios, enabling enhanced situational awareness, mission planning, and faster decision-making for both human operators and autonomous systems. We operate at the intersection of geospatial data, computer vision, and defense operations — and if that interests you, read on.
Responsibilities:
— Design, develop, and deploy production-ready computer vision algorithms for image and video analysis in GIS-related applications
— Build end-to-end perception pipelines — data collection, preprocessing, model development, deployment, and monitoring
— Integrate CV components into operational products in close collaboration with cross-functional engineering teams
— Optimize algorithms for performance, reliability, and scalability in real-time and resource-constrained environments
— Develop and deploy Visual Place Recognition (VPR) systems for geo-localization and scene understanding tasks
— Implement and optimize CV algorithms on embedded and hardware-accelerated platforms, ensuring stable and efficient inference at the edge
— Process and analyze large-scale datasets for model training, validation, and continuous improvement
— Diagnose and resolve complex issues across computer vision applications in production
— Contribute to internal research, exploring emerging techniques in image synthesis, manipulation, and applied ML.
Requirements:
— Expert-level Python — this is our primary engineering language. You must be comfortable writing clean, efficient, production-quality Python code
— 5+ years of hands-on experience in computer vision and applied machine learning
— Deep working knowledge of core CV and ML libraries: OpenCV, PyTorch, TensorFlow, and related tooling
— Solid understanding of machine learning, deep learning, and model development workflows end-to-end
— Hands-on experience with Visual Place Recognition (VPR) algorithms and techniques, including descriptor-based retrieval, sequence matching, and localization under varying conditions
— Proven ability to port, optimize, and run CV/ML models on embedded devices and edge hardware (e.g., NVIDIA Jetson, Raspberry Pi, or similar platforms)
— Proven experience with image processing, model training, and evaluation methodologies
— Ability to translate research into reliable engineering outputs with minimal supervision
Bachelor’s or Master’s degree in Computer Science, Artificial Intelligence, or a related field — or equivalent demonstrable experience
— Strong analytical thinking and clear communication in a cross-cultural, fast-moving team.
Would be a plus:
— Experience with graphics APIs: OpenGL, OpenGLES, or Direct3D
— Background in graphical algorithms, 3D rendering, or computational geometry
— Familiarity with embedded systems or hardware-accelerated proces