Senior Computer Vision Engineer (Action Recognition)
ScanditJob Description
📋 Description Define and prioritize experiments to validate CV model risks (precision, recall, latency) in retail. Design and implement end-to-end CV/ML pipelines on real-world video data. Set up and refine annotation workflow to create high-quality training/test datasets. Collaborate with product and engineering to clarify requirements and map to roadmaps. Select and optimize CV architectures for action recognition in unconstrained environments. Provide technical leadership in video CV, shaping practices and hiring. 🎯 Requirements University degree in CS, EE, Mathematics, or related field (advanced degree a plus). Hands-on CV/ML experience: design, train, evaluate, and deploy models in production. Proven track record building end-to-end CV/ML systems (0-1). Strong expertise in video-based pipelines and action/behavior recognition. Solid Python and ML/CV framework skills; work with real-world noisy data. Ability to work in dynamic, uncertainty-rich environments; prioritize experiments. Strong collaboration with cross-functional teams; clear English (B2+). Eligible to work in the hiring location. Nice to have: GDPR/privacy-aware data pipelines. Nice to have: annotation tools/workflows experience. Nice to have: retail or autonomous-store CV experience. 🎁 Benefits Flexible, office, hybrid or home working. Hackathons. Hybrid-friendly offices in Zurich, Warsaw, Tampere, London. Remote work support across Switzerland, Finland, Poland, UK, Italy and Germany. Attractive equity plan in a high-growth company. Birthday off; team outings; Great Place to Work certified.