Onur Copur
Computer Vision Engineer specialising in custom deep learning models, with expertise in object detection for retail and OCR solutions for postal services. Skilled in Python backend development in production environments, CI/CD automation, and hardware–software integration.
About Me
I'm a Computer Vision / AI Engineer with expertise in developing production-grade deep learning models. Currently working as AI Research Engineer at Imprima in Amsterdam, Netherlands, building agentic AI systems and scalable document processing pipelines for M&A workflows.
My work focuses on transformer-based OCR systems for national postal services and end-to-end microservice pipelines for retail shrinkage detection. I've successfully deployed systems handling 1M+ daily scans with <0.5% error rates and reduced retail losses by 25%.
I lead reading groups and technical seminars to keep teams at the forefront of AI research, and I'm passionate about modernizing legacy computer vision systems into scalable, production-ready solutions.
Technical Expertise
Research Interests
- OCR & Document Analysis
- Transformer Architectures
- Retail Computer Vision
- Edge AI & Deployment
- MLOps & Microservices
Hint: Try the classic cheat code... ↑↑↓↓←→←→BA