Curriculum Vitae

Onur Copur

Computer Vision / Deep Learning Engineer

onurcopur12@gmail.com
+31683518894
Amsterdam, Netherlands
in/onurcopur/
github.com/CopurOnur
Download PDF Resume

Professional Summary

Computer Vision Engineer specializing 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. Proficient in designing end-to-end data pipelines, deploying scalable AI solutions, and integrating models into edge devices.

Key Achievements

Transformer-Based OCR System for PostNL: Developed a novel transformer-based OCR achieving state-of-the-art accuracy and 65% lower latency, deployed in production for 1M+ daily scans with <0.5% error rate.
Real-Time Theft Prevention System for Albert Heijn: Built an end-to-end microservice pipeline for real-time shrinkage detection, reducing losses by 25%. Designed edge-optimized object detection models and embedding-based SKU matching across 50k+ items, deployed in 5 stores and 180 self-checkouts.
Luggage Tag Recognition and Matching System for Major Airports: Modernized OCR and object detection pipelines used in sorting systems at IST, AMS, OSL, DUB, and SIN airports, enabling robust recognition of diverse tag designs and high-throughput processing of thousands of luggage items daily under challenging imaging conditions.
Proof of Sorting for Austrian Post: Designed and deployed a computer vision verification system that automatically confirms whether parcels were sorted into the correct bins in real time. Built a low-latency image capture and analysis pipeline on edge devices, integrating object detection, tracking, and action recognition models to detect mis-sorted items and provide instant feedback to operators, improving sorting accuracy and operational efficiency across warehouse sites.

Professional Experience

AI Research Engineer

Imprima

Mar 2026 - Present
Amsterdam, Netherlands
  • Advancing redaction model performance by integrating state-of-the-art PII recognition models and fine-tuning them for mergers and acquisitions (M&A) document processing workflows.
  • Architecting scalable redaction and translation services leveraging Azure serverless GPU infrastructure to support high-throughput document pipelines.
  • Developing an agentic AI assistant embedded within the due diligence platform to streamline document review and accelerate decision-making workflows.

Computer Vision / Deep Learning Engineer

Prime Vision

May 2022 - Feb 2026
Delft, Netherlands
  • Developed and deployed production-grade deep learning models in containerized microservice architectures across multiple innovation projects, including PostNL, Albert Heijn, and Austrian Post.
  • Modernized legacy computer vision systems with transformer-based and edge-optimized deep learning models, achieving up to 65% latency reduction and measurable performance gains across logistics, warehouse management, aviation, and retail applications.
  • Led multiple proof-of-concept (PoC) studies to translate state-of-the-art computer vision research into robust, real-time solutions, including theft prevention, OCR modernization, and automated sorting verification.
  • Developed the 2026 technology adoption roadmap for the computer vision team, setting research priorities and guiding strategic initiatives to advance product capabilities.

Data Science Internship (Master’s Thesis - Computer Vision)

Barco

March 2021 - October 2021
Kortrijk, Belgium
  • Developed a deep learning system to analyze video data and assess subject engagement during educational content.
  • Implemented multi-task training approaches for engagement detection in e-learning environments.
  • Published research findings in ICIAP 2021 conference proceedings.

Data Scientist

Univera

May 2018 - August 2019
Izmir, Turkey
  • Developed a demand forecasting tool for retail, tobacco, and FMCG sectors.
  • Created a genetic algorithm route optimizer to improve logistics efficiency.
  • Built predictive models to optimize supply chain operations.

Education

Master's Degree in Data Science

Sapienza University

Graduated with High Honours (110 cum laude/110)

September 2019 - October 2021
Rome, Italy

Bachelor's Degree in Industrial Engineering

Yasar University

GPA: 3.77/4.0

September 2015 - June 2019
Izmir, Turkey

Areas of Expertise

Programming & Data Science

PythonSQLLangChainPandasopenai sdkPyTorchTensorFlowFastAPI

MLOps & Deployment

DockerKubernetesTritonMLFlowTeamCityAirflowONNXRedisPinecone

Cloud & Infrastructure

Microsoft AzureAWSLinux

Edge AI & Embedded CV

Nvidia JetsonOAK-D CamerasESP32

Project Management

SCRUMJiraCI/CDGit

Selected Publications

"Renewable energy management in smart home environment via forecast embedded scheduling based on Recurrent Trend Predictive Neural Network"

O. Copur et al.Applied Energy, 340 (2023)

"Engagement detection with multi-task training in e-learning environments"

O. Copur et al.ICIAP (2021)

For complete publication details, see my Google Scholar profile

Languages

Turkish (Native)English (C2)Dutch (B1)