Once Sport is a sports video analysis software company seeking a Computer Vision Engineer to join its team.
The Once Sport product suite includes the Once Sport Analyser, Coach Board, and Autocam. The Analyser provides advanced tagging, animated 3D tracking graphics, and full control over live analysis, enabling coaches across multiple sports worldwide. Coach Board enables creation of animated training sessions and tactical scenarios on a 2D tactical board, while Autocam autonomously tracks the game and generates TV-quality footage without a camera operator.
With clients worldwide, including clubs and coaches in the Bundesliga, UEFA Champions League, Croatian Football League, and international football federations, we are expanding into new markets such as the United States.
This is an opportunity for a computer vision engineer to develop and optimise production systems, work with large-scale sports video data, and directly influence how elite-level football is analysed.
Learn more at https://once.sport
More about the position
As a Computer Vision Engineer, you will research and develop ML-based approaches for understanding football video and detecting key match events such as goals, shots, fouls and corners. Your work will focus on analysing full-game footage, localising when important moments happen, and turning them into useful insights for coaches, analysts and players. You will explore state-of-the-art video understanding and multimodal models that can learn directly from match footage, helping us reduce reliance on handwritten, rule-based systems. This role includes experimenting with new architectures, building training and evaluation pipelines, and bringing successful research ideas into real-world Once Sport products.
Our development environment, tools & technologies
Our software works on all major operating systems as well as in the cloud.
- Pytorch & ONNX Runtime for neural networks
- OpenCV for traditional CV tasks
- C/C++ for low-level image operations
- Avalonia UI framework and Skia for the GUI
- FFmpeg for handling video
- Golang for the backend
What we offer:
- Hybrid work model
- A flexible schedule and relaxed atmosphere that fosters creativity and productivity
- Exposure to the latest technologies and trends in AI, video analysis, and sports
How to apply?
Send your CV at dalibor@once.sport
We can’t wait to hear from you!
