Kujtim is available for hire
Hire KujtimKujtim Rahmani
Verified Expert in Engineering
Machine Learning Developer
Kumanovo, Municipality of Kumanovo, Macedonia
Toptal member since August 7, 2023
Bio
Kujtim is a computer vision and machine learning engineer with over ten years of experience and more than four years of experience in academia. During his tenure, he worked with four startups, where he built products from scratch. Furthermore, Kujtim also worked as a consultant for esteemed clients such as the German automotive industry and Heraeus Group and as a data scientist researcher for Airbus.
Portfolio
Klues
Computer Vision, Python, You Only Look Once (YOLO), Computer Vision Algorithms...
Semos Education
Optical Character Recognition (OCR), Machine Learning, OpenAI...
Jack Zerby Consulting, LLC
Artificial Intelligence (AI), Machine Learning, Computer Vision, Python...
Experience
- OpenCV - 9 years
- Computer Vision - 9 years
- Python - 9 years
- Deep Learning - 5 years
- PyTorch - 4 years
- You Only Look Once (YOLO) - 4 years
- Intelligent Character Recognition - 4 years
- Defect Management - 4 years
Availability
Part-time
Preferred Environment
Ubuntu, Windows, Visual Studio Code (VS Code)
The most amazing...
...thing I've developed is the SOTA algorithm for defect detection for airplane parts. I used traditional ML to build SOTA for segmentation in a small dataset.
Work Experience
Computer Vision and YOLO Developer
2024 - PRESENT
Klues
- Advised on the dataset acquisition and model selection strategies for problems.
- Created the AI infrastructure to develop the detection and counting problem in urban scenes.
- Built the AI architecture for different use cases for outdoor scenes, civil engineering, and GIS data.
Technologies: Computer Vision, Python, You Only Look Once (YOLO), Computer Vision Algorithms, Artificial Intelligence (AI), AI Algorithms, AI Consulting, AI Model Training, Algorithm Design, System Design, Solution Design
Trainer and AI Project Architect
2023 - PRESENT
Semos Education
- Mentored students and contributed to industrial projects in the fintech and manufacturing sectors.
- Trained banks, financial institutions, and manufacturing factories in using artificial intelligence.
- Integrated a generative AI solution to a large bank.
Technologies: Optical Character Recognition (OCR), Machine Learning, OpenAI, Azure OpenAI Service, Azure, Computer Vision, PyTorch, Convolutional Neural Networks (CNNs), Object Recognition, Optimization, Machine Learning Algorithms, PDF, AI Algorithms, Hugging Face, Fine-tuning, Transformers, AI Model Training, Algorithm Design, Data Cleansing, Mathematical Modeling, Image Classification, Cloud
AI/ML Expert and Consultant
2023 - PRESENT
Jack Zerby Consulting, LLC
- Created an algorithm for the intelligent processing of collectible images. Developed several supervised detection and intelligent character processing of collectibles. Handled several templates.
- Helped the team integrate the service in the AWS and contributed to designing the app.
- Fine-tuned several deep learning models to process and better recognize the text in the images, as well as handle the errors made by the recognizer.
Technologies: Artificial Intelligence (AI), Machine Learning, Computer Vision, Python, Image Recognition, Convolutional Neural Networks (CNNs), Optical Character Recognition (OCR), Information Extraction, OpenCV, Deep Learning, Image Processing, Intelligent Character Recognition, Pytesseract, Tesseract, You Only Look Once (YOLO), Data Structures, Object Detection, AI Consulting, PyTorch, Object Recognition, Machine Learning Algorithms, PDF, AI Algorithms, Ubuntu, Fine-tuning, AI Model Training, Algorithm Design, Data Cleansing, Feature Engineering, Mathematical Modeling, Image Classification, Cloud
Senior Data Scientist | Computer Vision
2023 - PRESENT
Pano
- Built computer vision algorithms for fire detection and tracking in forests.
- Developed mathematical modeling of the fire initiation and growth during the night. Used Python, OpenCV, scikit-learn, GCP, and GitHub.
- Created a tracking algorithm for tracking forest fires during the night.
Technologies: Python, PyTorch, OpenCV, Machine Learning, Computer Vision, Artificial Intelligence (AI), Data Science, Matplotlib, Deep Learning, Image Processing, Algorithms, Object Tracking, Object Detection, You Only Look Once (YOLO), Probability Theory, Statistics, Convolutional Neural Networks (CNNs), Loss Modeling, NumPy, Computer Vision Algorithms, Visualization, Video Processing, Neural Networks, Startups, Data Structures, Video Analysis, AI Consulting, Object Recognition, Machine Vision, Optimization, Machine Learning Algorithms, AI Algorithms, Ubuntu, Hugging Face, Fine-tuning, Transformers, Image Segmentation, AI Model Training, Algorithm Design, Data Cleansing, Feature Engineering, Mathematical Modeling, Image Classification, Cloud, Video Analytics
Senior Computer Vision Engineer
2021 - PRESENT
Rare Edition
- Built computer vision algorithms for card and coin defect evaluation. Conducted quality assurance for collectibles.
- Communicated with operations (graders), back-end services, and the mobile developer to ensure all computer vision algorithms ran correctly.
- Created an active learning platform for card grading.
Technologies: Python, Amazon Web Services (AWS), TensorFlow, Keras, Docker, SQL, Machine Learning, Computer Vision, Artificial Intelligence (AI), Data Analysis, Amazon S3 (AWS S3), Matplotlib, Deep Learning, Image Processing, Algorithms, Linear Regression, PostgreSQL, Object Detection, Pandas, You Only Look Once (YOLO), Image Recognition, Probability Theory, Statistics, PySQL, Convolutional Neural Networks (CNNs), Data Analytics, NumPy, Computer Vision Algorithms, Data Structures, Database Design, Databases, Visualization, Labeling, Neural Networks, Startups, Intelligent Character Recognition, Information Extraction, Dynamic Programming, OpenCV, Defect Management, AI Consulting, Object Recognition, Machine Vision, Optimization, Machine Learning Algorithms, AI Algorithms, Ubuntu, TensorBoard, Fine-tuning, Quantization, Image Segmentation, AI Model Training, Algorithm Design, Data Cleansing, Feature Engineering, Mathematical Modeling, Image Classification, Cloud
Staff Engineer
2025 - 2025
Gatik
- Integrated tfRecord reading component for multi-node training.
- Integrated MLFlow visualization and Azure ML MLflow.
- Trained using Ray.io and interviewed candidates for a research scientist role in Python, machine learning, and algorithms.
Technologies: Artificial Intelligence (AI), Machine Learning, Fine-tuning, Transformers, LSTM Networks, Convolutional Neural Networks (CNNs), Quantization, Large Language Models (LLMs), NVIDIA TensorRT, NVIDIA CUDA, Kubeflow, Argo CD, Google Cloud Functions, Tekton, Jenkins, GitHub Actions, Horovod, PyTorch, Ray Train, TensorFlow, Dask, Kubernetes, Video Analytics
Machine/Computer Vision Engineer/Consultant
2024 - 2025
Hidonix Inc
- Built an AI unsupervised algorithm for museum object detection, segmentation, and identification, as well as a search engine for museum objects.
- Advised on the AI architecture of the detection and recognition system.
- Built an online and offline search algorithm for specific museums.
Technologies: Object Recognition, Image Recognition, Python, PyTorch, OpenCV, Computer Vision, Machine Vision, Convolutional Neural Networks (CNNs), TensorFlow, FAISS, Unsupervised Learning
Senior Computer Vision Solution Architect
2024 - 2024
Peggy
- Constructed an algorithm for false Artwork detection.
- Constructed a fast CPU feature extractor and matcher to identify non-original artworks that were completed in less than two seconds.
- Constructed an algorithm for classifying artworks in several categories and the quality of the artwork.
- Developed a LightGlue algorithm and statistical feature extraction to analyze the artworks in real-time.
Technologies: Feature Analysis, PyTorch, Flutter, Amazon Web Services (AWS), Dart, Augmented Reality (AR), Pandas, Geometry, Epipolar Geometry, Statistical Modeling
Senior AI/Computer Vision Engineer
2024 - 2024
peggy
- Constructed an algorithm for false artwork detection.
- Constructed a fast CPU feature extractor and matcher to identify non-original artworks that were completed in less than two seconds.
- Constructed an algorithm for classifying artworks into several categories and the quality of the artwork.
- Developed the LightGlue algorithm and statistical feature extraction to analyze the artworks in real time.
Technologies: OpenCV, Python, Machine Learning, Feature Analysis, Image Processing, Pandas, Flutter, C++, AI Model Training, Algorithm Design, Data Cleansing, Feature Engineering, Mathematical Modeling, Image Classification, Cloud
Machine Learning Engineer
2023 - 2023
CollX
- Developed an image search and matching engine for sports like baseball, basketball, and soccer and game cards such as Pokemon and Magic. The users can take a photo of their card, and my algorithm finds the matching card in the system.
- Used contrastive and triplet loss metrics to build an image embedding system.
- Created a hard example mining system for algorithm improvement.
- Built a Faiss indexer to index the card images and integrated feature matching to filter the best results.
Technologies: PyTorch, FAISS, OpenCV, Open Neural Network Exchange (ONNX), Flask, Pandas, Machine Learning, Computer Vision, Artificial Intelligence (AI), Data Analysis, Matplotlib, Deep Learning, Image Processing, Data Engineering, Algorithms, PostgreSQL, Image Search, Image Recognition, Probability Theory, Statistics, Convolutional Neural Networks (CNNs), Loss Modeling, Data Analytics, NumPy, Databases, Visualization, Labeling, Neural Networks, Startups, Intelligent Character Recognition, Information Extraction, Python, AI Consulting, Machine Vision, Optimization, Machine Learning Algorithms, AI Algorithms, Ubuntu, Fine-tuning, Quantization, AI Model Training, Algorithm Design, Data Cleansing, Feature Engineering, Image Classification, Cloud
Senior Data Scientist | Computer Vision
2023 - 2023
Ramani
- Built computer vision algorithms for object detection, counting, and tracking in warehouses.
- Conducted 3D reconstruction of the objects based on multiple images.
- Managed junior computer vision engineers and annotators.
Technologies: PyTorch, Python, OpenCV, Scikit-learn, Amazon Web Services (AWS), Graph Theory, Machine Learning, Computer Vision, Artificial Intelligence (AI), Data Science, Matplotlib, Deep Learning, Image Processing, Data Engineering, 3D Image Processing, 3D Reconstruction, Object Detection, Flask, Probability Theory, Statistics, Convolutional Neural Networks (CNNs), Open Neural Network Exchange (ONNX), NumPy, Computer Vision Algorithms, Data Structures, Visualization, Labeling, Neural Networks, Startups, Dynamic Programming, Intelligent Character Recognition, Pytesseract, Optical Character Recognition (OCR), You Only Look Once (YOLO), AI Consulting, Object Recognition, Machine Vision, Optimization, Machine Learning Algorithms, AI Algorithms, Ubuntu, Fine-tuning, Image Segmentation, AI Model Training, Algorithm Design, Data Cleansing, Feature Engineering, Mathematical Modeling, Cloud
Senior Data Scientist | Computer Vision
2023 - 2023
Smartauger
- Built computer vision algorithms for objects like pipes and optical fiber cables and their detection in electromagnetic wave-generated images from ground-penetrating radars.
- Ran experiments and built the algorithm using simulated and real data.
- Guided the stakeholders on how to acquire the data from real scenarios.
Technologies: PyTorch, Computer Vision, Amazon Web Services (AWS), Scikit-learn, Object Detection, Machine Learning, Artificial Intelligence (AI), Data Analysis, Amazon S3 (AWS S3), Matplotlib, Deep Learning, You Only Look Once (YOLO), Probability Theory, Statistics, Convolutional Neural Networks (CNNs), Data Analytics, NumPy, Computer Vision Algorithms, Labeling, Neural Networks, Startups, OpenCV, Image Processing, Tesseract, Defect Management, AI Consulting, Object Recognition, Machine Vision, Machine Learning Algorithms, AI Algorithms, Ubuntu, Fine-tuning, AI Model Training, Algorithm Design, Data Cleansing, Cloud
Senior Data Scientist | Computer Vision
2022 - 2023
Heraeus
- Built computer vision algorithms for object detection and tracking of defects of manufactured parts in X-ray images.
- Optimized the detection and tracking to process more than 17 images per second.
- Collaborated with the clients to ensure the data and annotation were of the best quality.
Technologies: Python, PyTorch, OpenCV, Open Neural Network Exchange (ONNX), You Only Look Once (YOLO), Kalman Filtering, Scikit-learn, Random Forests, Pandas, Machine Learning, Computer Vision, Artificial Intelligence (AI), Data Science, Azure, Matplotlib, Deep Learning, Image Processing, Decision Trees, Data Engineering, Algorithms, Object Tracking, Object Detection, Probability Theory, Statistics, Convolutional Neural Networks (CNNs), NumPy, Computer Vision Algorithms, Visualization, Video Processing, Labeling, Neural Networks, Defect Management, Video Analysis, AI Consulting, Object Recognition, Machine Vision, Optimization, Machine Learning Algorithms, AI Algorithms, Fine-tuning, AI Model Training, Algorithm Design, Data Cleansing, Feature Engineering, Mathematical Modeling, Image Classification, Cloud, Video Analytics
Machine Learning Engineer
2022 - 2022
Tolstoy
- Built an OCR automatic parsing of the electricity bills.
- Integrated Google Drive with the Cloud Run and Cloud Scheduler to ensure new data was parsed after being uploaded to Google Drive.
- Handled several electricity bill templates and file formats.
Technologies: Tesseract, Pytesseract, Optical Character Recognition (OCR), Docker, Google Cloud Platform (GCP), Pandas, Flask, Machine Learning, Artificial Intelligence (AI), Data Analysis, Data Engineering, Probability Theory, Statistics, NumPy, Databases, Startups, Intelligent Character Recognition, Information Extraction, OpenCV, Python, AI Consulting, PyTorch, Optimization, Machine Learning Algorithms, PDF, AI Algorithms, Ubuntu, Fine-tuning, AI Model Training, Algorithm Design, Data Cleansing, Feature Engineering, Image Classification, Cloud
Data Scientist Consultant
2019 - 2021
Cluster Reply
- Built an object detection and tracking algorithm for counting objects in 3D computed tomography images of vehicle parts. Reduced the time from around 120 minutes to five minutes.
- Created an invoice data extraction algorithm for a private German digital post service.
- Developed NLP algorithms for a research paper, understanding and computing its correlation to each of the UN countries' sustainable development goals.
- Prepared machine learning and data science project proposals for customers.
Technologies: Python, OpenCV, Keras, TensorFlow, Azure Cognitive Services, Scikit-learn, Kalman Filtering, Machine Learning, Computer Vision, Artificial Intelligence (AI), Data Science, Data Analysis, Azure, Matplotlib, Scikit-image, Deep Learning, Image Processing, Decision Trees, Data Engineering, Random Forests, Algorithms, Linear Regression, Object Tracking, 3D Image Processing, Object Detection, Deep Sorting, Probability Theory, Statistics, PySQL, Convolutional Neural Networks (CNNs), Data Analytics, NumPy, Computer Vision Algorithms, Data Structures, Databases, Visualization, Video Processing, Labeling, Neural Networks, Intelligent Character Recognition, Information Extraction, Dynamic Programming, Pytesseract, You Only Look Once (YOLO), Defect Management, Azure Functions, Video Analysis, AI Consulting, Object Recognition, Optimization, Machine Learning Algorithms, PDF, AI Algorithms, TensorBoard, Fine-tuning, Transformers, LSTM Networks, Image Segmentation, AI Model Training, Algorithm Design, Data Cleansing, Feature Engineering, Mathematical Modeling, Image Classification, Cloud, Video Analytics
Data Scientist
2018 - 2019
Airbus
- Built a pore detection algorithm in 3D CT scans of airplane parts with 99.9% accuracy.
- Created a system design for change detection in airplane cabins using radio frequency signals.
- Guided the material scientist on data annotation and understanding how ML algorithms work.
Technologies: Python, TensorFlow, Scikit-learn, Amazon Web Services (AWS), Machine Learning, Computer Vision, Artificial Intelligence (AI), Data Science, Amazon S3 (AWS S3), Matplotlib, Image Processing, Algorithms, Linear Regression, 3D Image Processing, SVMs, Probability Theory, Statistics, Data Analytics, NumPy, Computer Vision Algorithms, Research, Visualization, Labeling, Neural Networks, OpenCV, Object Detection, Defect Management, Video Analysis, Object Recognition, Optimization, Machine Learning Algorithms, AI Algorithms, TensorBoard, Fine-tuning, Image Segmentation, AI Model Training, Algorithm Design, Data Cleansing, Feature Engineering, Mathematical Modeling, Image Classification, Video Analytics
Research Assistant and Scientist
2015 - 2018
UniBW
- Developed a semantic scene segmentation algorithm for facade images, which is state of the art for small datasets.
- Created a dataset for detailed facade segmentation, including transom and mullion windows. Analyzed and considered data for creating high-quality 3D building models.
- Built a complete pipeline for facade segmentation from data acquisition up to deployment.
Technologies: C++, OpenCV, TensorFlow, Keras, Machine Learning, Computer Vision, Artificial Intelligence (AI), Data Science, Matplotlib, Deep Learning, Decision Trees, Data Engineering, Random Forests, Algorithms, Object Detection, Number Theory, Probability Theory, Statistics, Convolutional Neural Networks (CNNs), Loss Modeling, Data Analytics, NumPy, Computer Vision Algorithms, Research, Data Structures, Visualization, Labeling, Neural Networks, Dynamic Programming, Image Processing, Python, Object Recognition, Optimization, Machine Learning Algorithms, AI Algorithms, Ubuntu, TensorBoard, Fine-tuning, Image Segmentation, AI Model Training, Algorithm Design, Data Cleansing, Feature Engineering, Mathematical Modeling
Machine Learning Engineer
2014 - 2015
Zyncd
- Built a matching algorithm to match user profiles with problem descriptions.
- Improved the accuracy of the matching by more than 50%.
- Developed a ranking algorithm using mathematical functions to normalize profiles.
- Integrated word2vec into the system for profile embedding.
Technologies: Python, Scikit-learn, CherryPy, Machine Learning, Artificial Intelligence (AI), Data Science, Data Engineering, Algorithms, Probability Theory, Statistics, Data Analytics, NumPy, Database Design, Databases, Neural Networks, Startups, Information Extraction, Dynamic Programming, Deep Learning, Web Development, Optimization, Machine Learning Algorithms, AI Algorithms, Ubuntu, Fine-tuning, AI Model Training, Algorithm Design, Data Cleansing, Feature Engineering
Computer Lab Teacher
2011 - 2012
Ss. Cyril and Methodius University in Skopje
- Held calculus courses for first and third-year students.
- Conducted discrete mathematics courses for first-year students.
- Taught a course on the basics of IT for first-year students.
Technologies: Mathematics, Algorithms, Number Theory, Statistics, Probability Theory, Dynamic Programming, AI Algorithms, Algorithm Design
Experience
Image-searching Engine for Cards
http://collx.appDeveloped an image-matching algorithm for searching cards. The database contains millions of images. The user can take a photo, and the algorithm will find the cards that are the same as the photo and show the cards and additional information about them to the user.
Pore Detection Algorithm in 3D CT Scans
A pore detection algorithm in 3D CT scans of airplane parts with 99.9% accuracy. I built the algorithm using several Gaussian modifications and transformed data to an n-modal normal distribution. Furthermore, I made high-quality decision boundaries between all n distributions and classified each region on one of these distributions. Also, I worked on simulating the data.
Object Detection in 3D-computed Tomography Images
An object detection and tracking algorithm that I built for counting objects in 3D CT images of vehicle parts. Reduced the human (worker) time from around 120 minutes to 5 minutes.
Each CT image was around 1GB. Since there was no pre-trained model for 3D images, I converted the problem to object detection in each slice of the 3D image and then used tracking and dynamic programming to recreate and delineate the 3D objects.
Each CT image was around 1GB. Since there was no pre-trained model for 3D images, I converted the problem to object detection in each slice of the 3D image and then used tracking and dynamic programming to recreate and delineate the 3D objects.
Semantic Scene Segmentation Algorithm for Façade Images
https://isprs-annals.copernicus.org/articles/IV-2/223/2018/isprs-annals-IV-2-223-2018.pdfA semantic segmentation pipeline that I built for façades. This is a hybrid pipeline consisting of traditional machine learning algorithms (the structured random forest model), an iterative optimization technique, a deep learning region proposal network, and a dynamic programming optimization technique based on the façade object constraints.
Information Extraction from Collectibles
I built information extraction, character recognition, and information structuring algorithms for collectibles. My role was to make the brain (AI) of the application. Furthermore, I was constantly communicating with the developers and the stakeholders since I was the person who knew both the market and the technology.
Education
2016 - 2019
PhD in Artificial Intelligence
University of the Bundeswehr Munich - Munich, Germany
2012 - 2014
Master's Degree in Computer Science
Technical University of Munich - Munich, Germany
2008 - 2012
Bachelor's Degree in Computer Science
Ss. Cyril and Methodius University in Skopje - Skopje, Macedonia
Certifications
MARCH 2024 - PRESENT
Certified Microsoft Trainer
Microsoft
MARCH 2024 - MARCH 2025
Azure AI Associate
Microsoft
JANUARY 2024 - PRESENT
CertNexus Certified Artificial Intelligence Practitioner (CAIP)
CertNexus
JANUARY 2024 - JANUARY 2027
CertNexus Certified Artificial Intelligence Practitioner (CAIP)
CertNexus
OCTOBER 2020 - OCTOBER 2024
Microsoft Certified: Azure Data Scientist Associate
Microsoft
Skills
Libraries/APIs
OpenCV, PyTorch, NumPy, TensorFlow, Keras, Scikit-learn, Matplotlib, Pandas, Azure Cognitive Services, Ray Train, Dask
Tools
You Only Look Once (YOLO), Algorithm Design, Open Neural Network Exchange (ONNX), TensorBoard, Scikit-image, IPython Notebook, ChatGPT, Azure OpenAI Service, Jenkins
Languages
Python, C++, SQL, Dart
Paradigms
Dynamic Programming, Database Design
Platforms
Ubuntu, Amazon Web Services (AWS), Azure, Docker, Google Cloud Platform (GCP), Azure Functions, NVIDIA CUDA, Kubeflow, Kubernetes
Storage
Amazon S3 (AWS S3), PostgreSQL, Databases
Frameworks
Flask, CherryPy, Flutter, Horovod
Industry Expertise
Teaching
Other
Machine Learning, Deep Learning, Computer Vision, Image Processing, Random Forests, Mathematics, Algorithms, Data Structures, Object Detection, Artificial Intelligence (AI), Data Science, Data Analysis, Decision Trees, Probability Theory, Statistics, Convolutional Neural Networks (CNNs), Data Analytics, Computer Vision Algorithms, Research, Labeling, Neural Networks, Startups, Intelligent Character Recognition, Defect Management, Video Analysis, AI Consulting, Object Recognition, Optimization, Machine Learning Algorithms, AI Algorithms, Fine-tuning, Transformers, Image Segmentation, AI Model Training, Data Cleansing, Feature Engineering, Mathematical Modeling, Image Classification, Video Analytics, Natural Language Processing (NLP), Kalman Filtering, Optical Character Recognition (OCR), Tesseract, Data Engineering, Linear Regression, Object Tracking, 3D Image Processing, Image Search, Image Recognition, Number Theory, PySQL, Loss Modeling, Visualization, Video Processing, Information Extraction, Machine Vision, Large Language Models (LLMs), PDF, Hugging Face, Quantization, Cloud, Graph Theory, FAISS, Pytesseract, 3D Reconstruction, SVMs, Deep Sorting, Statistical Methods, Statistical Modeling, OpenAI, Training, Web Development, AI Chatbots, LSTM Networks, Feature Analysis, NVIDIA TensorRT, Argo CD, Google Cloud Functions, Tekton, GitHub Actions, Unsupervised Learning, System Design, Solution Design, Augmented Reality (AR), Geometry, Epipolar Geometry
Collaboration That Works
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
1
Share your needs
Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2
Choose your talent
Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3
Start your risk-free talent trial
Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.
Top talent is in high demand.
Start hiring