We are looking for a developer to build a football video analysis tool that can track players, detect events, and generate detailed match data. Core Requirements: • Player detection in match videos using YOLOv8 (or similar) • Player tracking with Deep SORT or similar tracking algorithms • Event recognition: passes, shots, pressures, duels, recoveries, off-ball movements • Generation of heatmaps, movement maps, and structured match reports • Process full match videos (90 minutes) offline • Modular design to allow adding more statistics/features later Preferred Tech Stack: • Computer Vision: OpenCV, YOLOv8, DeepSORT • Machine Learning: TensorFlow / PyTorch • Data Handling: Python, SQL/NoSQL • Optional: Docker for deployment, cloud GPU experience Questions for Applicants: • Relevant previous experience in tracking/detection/sports analytics • Approach to handling event recognition with limited labeled data • Ideas for structuring the system for scalability