Senior Computer Vision & ML Engineer – Action Recognition (Hybrid)

Scandit; Warsaw (IN); Hybrid; USD 46,000 - 77,000; Full time

Benefits offered by this job

Flexible, office, hybrid or home work

Your birthday off

Great Place to Work in 7 countries

Job summary

Scandit Warsaw is seeking a Senior Computer Vision/Machine Learning Engineer to lead the technical discovery and early development of a new video/action recognition solution for retail. You will own end-to-end pipelines, validate risk assumptions, and deliver robust, privacy-aware CV models for live environments.

You will collaborate with product and engineering to turn real-world video data into production-ready systems, applying state-of-the-art architectures and practical constraints in a

Qualifications

• Strong background in computer vision and machine learning with production experience.
• Experience designing, training, evaluating and deploying CV/ML models in real systems.
• Proven ability to ship end-to-end video analysis solutions at scale.

Responsibilities

• Lead technical discovery and early development of new CV/ML solutions for retail video data.
• Design end-to-end pipelines from data ingestion to model deployment in production.
• Collaborate with product and engineering to translate requirements into roadmaps.
• Develop privacy-aware, robust vision systems for live environments.
• Mentor teams and influence hiring and best practices in CV/ML.

Skills

Computer vision

Machine learning

Python

English (B2)

Education

University degree in Computer Science / Electrical Engineering / Mathematics

Tools

PyTorch

TensorFlow

CV frameworks

Job description

Scandit Warsaw is seeking a Senior Computer Vision/Machine Learning Engineer to lead the technical discovery and early development of a new video/action recognition solution for retail. You will own end-to-end pipelines, validate risk assumptions, and deliver robust, privacy-aware CV models for live environments.

You will collaborate with product and engineering to turn real-world video data into production-ready systems, applying state-of-the-art architectures and practical constraints in a