Nenad Biresev-Freelance Computer Vision Engineer
Check rate
Experience
Safety Video Analytics Project for Airbus
Airbus
- Developed a real-time video analytics proof-of-concept for deployment on NVIDIA Jetson edge devices.
- Implemented DeepStream pipelines including object detection, tracking, human pose estimation, face anonymization, and zone intrusion detection.
- Built a Qt/Python demonstration UI interfacing with the AI pipeline via REST APIs.
Autonomous Driving Data Quality — DB InfraGO
DB InfraGO
- Developed Python pipelines to analyze multi sensor datasets recorded as ROS2 bag files.
- Processed synchronized streams from Radar, LiDAR, RGB, Thermal, and ToF cameras used in autonomous train perception systems.
- Implemented automated quality checks detecting corrupted bags, missing messages, synchronization issues, and incorrect formats.
Freelance Computer Vision Engineer
Vorwerk & Co. KG
- Improved obstacle detection performance for autonomous cleaning robots by ~12% through dataset optimization and failure-case analysis.
- Developed multi-modal perception pipelines combining ToF depth and intensity data with YOLO detection and semantic segmentation.
- Deployed optimized C++ inference pipelines on ARM64 embedded robotic platforms.
- Implemented map-based robot intelligence including no-go zones and optimized cleaning routes using A* and Traveling Salesman optimization.
Freelance Computer Vision Engineer
Veridos GmbH
- Developed multi-modal biometric anti-spoofing systems for airport automated border control.
- Built RGB + thermal face detection and presentation attack detection pipelines improving spoof detection accuracy by ~8% (~99.2% overall).
- Developed DeepLabV3+ portrait background removal achieving ~98.8% mIoU.
- Contributed to document intelligence including passport damage detection and OCR pipelines.
Live Casino Computer Vision
Pragmatic Play
- Developed real-time computer vision systems for live casino games (roulette and blackjack) using live camera feeds deployed at multiple casino sites.
- Built detection pipelines for roulette ball tracking, number recognition, ball landing position, and rotation counting; and blackjack card recognition with continuous game state monitoring.
- Engineered a high-throughput inference pipeline achieving 200 FPS on NVIDIA Jetson edge devices by adapting a MobileNetV2-SSD architecture with a simplified backbone and head to meet strict latency constraints.
- Integrated results upload via REST API to Azure-hosted databases and deployed training pipelines on Azure compute clusters.
- Implemented production monitoring with Prometheus metrics and alerting, and maintained a CI/CD pipeline with Jenkins.
Machine Learning Engineer for RPA
Freelance
- Conducted R&D into autonomous game-playing agents for card game environments, implementing real-time card detection, hand recognition, and decision-making logic across multiple edge devices.
- Developed deep learning pipelines for object detection and UI element classification using TensorFlow
- Implemented automation and RPA solutions using Python and C++.
- Deployed applications on-premises and in Azure cloud environments
Machine Learning Engineer - Computer Vision
Motec GmbH
- Trained deep learning models for object detection tasks to recognize humans, cars, trees and other objects in RGB images using Python and TensorFlow
- Implemented ground plane detection and road lane detection using the ZED stereo camera with classical computer vision approaches in C++ and OpenCV
Software Developer
Gigatronik Holding GmbH
- Developed a face recognition iOS app using Xcode, Objective C and OpenCV allowing users to register accounts and login via face recognition
- Implemented gesture recognition on Jetson with Asus Xtion Pro Camera using C++, OpenCV, OpenNI and Nite to record and recognize hand gestures as input commands
- Built a Ford Sync iOS app to control Ford Sync via Bluetooth Low Energy using Xcode and Objective C
- Created a peripheral and central simulator for BLE devices in an iOS app using Xcode and Objective C
- Developed a Mac OS X driver for a Bosch eBike device in C and C++
- Implemented HMI for BHTC Multi-Touch demonstrator in Visual Studio 2013 with C#, Windows Forms and WPF providing handwriting recognition, image manipulation and finger gesture recognition
- Developed HMI for BHTC demonstrator controlling air conditioner in a car using C++, Qt and QML
- Delivered EVOBUS project for Mercedes-Benz - Daimler AG implementing HMI for an autonomous driving bus using WPF, C#, MVVM and Visual Studio 2015
- Implemented multi-language System Xentry tabs and diagnosis systems for Mercedes-Benz - Daimler AG using WPF, C#, MVVM and Visual Studio 2015
- Created HMI multimedia demonstrator for Neighborhood Electric Vehicles from Gigatronik using C++, Qt and QML in a SCRUM environment
Scientific Researcher
University of Bonn
- Researched hierarchical invariant features for action recognition in videos using C++ and Python
- Accelerated algorithms with CUDA C++ extensions
Software Developer
Fraunhofer-Institute for Intelligent Analyses
- Performed object class segmentation using random forests and RGB-D SLAM
- Built semantic maps using random forests for object class segmentation
- Contributed to the Nifti rescue robot project
Software Developer
Ipsylon
- Developed Windows Form applications in Visual Studio 2008 using C#, SQL Server 2005 and LINQ to SQL
Industry Experience
See where this freelancer has spent most of their professional time.
Experienced in Information Technology, Government and Administration, Automotive, Manufacturing, Education, and Aerospace and Defense.
Business Area Experience
See which departments and functions this freelancer has contributed to most.
Experienced in Product Development, Research and Development, Information Technology, and Quality Assurance.
Summary
Computer Vision Engineer with 10+ years specializing in AI driven visual perception and 15+ years in software development with a track record of taking systems from research prototype to production hardware. Built and deployed real world perception pipelines across autonomous robotics, live biometric border control, safety critical drowning detection, and high throughput video analytics at 200 FPS on edge devices. Comfortable at every layer of the stack from CUDA optimized model architectures on NVIDIA Jetson to cloud training pipelines on Azure and Databricks. Driven by hard constraints: latency, reliability, and accuracy that holds up outside the lab.
Skills
- C++: 6 Years
- Python: 8 Years
- C#: 7 Years
- Opencv: 8 Years
- Tensorflow: 4 Years
- Pytorch: 4 Year
- Qt: 5 Years
- Wpf: 5 Years
- Mlflow: 2 Years
- Dvc: 1 Year
- Lakefs: 1 Year
- Docker: 4 Years
- Flyte: 2 Year
- Azure Cloud: 4 Years
- Google Cloud 6 Months
- Os: Windows, Linux, Os X
- Databricks: 2 Years
- Aws: 2 Years
Languages
Education
Rheinische Friedrich-Wilhelms University - Bonn
Master of Science, Computer Science · Computer Science · Bonn, Germany
Institute For Mathematics Und Computer Science - Novi Sad, Serbia
Master of Science, Computer Science · Computer Science · Novi Sad, Serbia · 9.1/10.00
Statistics
Experience
Global Experience
Expertise
Qualifications
Profile
Frequently asked questions
Have questions? Find more information here.
Average rates for similar positions
Rates are based on recent contracts and do not include FRATCH margin.
Similar Freelancers
Discover other experts with similar qualifications and experience
Experts recently working on similar projects
Freelancers with hands-on experience in comparable project as a Safety Video Analytics Project for Airbus
Nearby freelancers
Professionals working in or nearby Bonn, Germany
