Computer Vision Engineer
HybridLondon, England, United Kingdom
Job Summary
Build computer vision models and pipelines that extract key data points from cricket broadcasts and replays, handling high-variation sources like inconsistent framing and broadcast-quality footage. Develop a robust framework to validate automatically extracted data against manually collected ground-truth, ensuring events are not double-counted or falsely captured. Scale systems to process large volumes of historical footage with consistent quality while identifying opportunities to increase automation and reduce manual effort. Lead technical decisions and define best practices for the CV team, collaborating with Quant researchers and domain experts to automate the entire data pipeline.
Required Qualifications
- Strong experience building and deploying computer vision models in production
- Expertise in areas such as object detection, tracking, pose estimation, action recognition, or temporal event detection
- Confidence working autonomously and leading and owning technical decisions
- Ability to collaborate with Quant researchers, cricket domain experts, and data engineers
- A pragmatic mindset—balancing research exploration with delivery
Hiring someone like this?
Get your role in front of qualified candidates on Sorce.