AHMAD PAHRURROZI

Key Skills

Software

Roboflow

Top Subject Matter

No subject matter listed

Top Data Types

Image

Top Label Types

Bounding Box

Polygon

Freelancer Overview

I am a dedicated computer vision data labeller with hands-on experience preparing and annotating high-quality image datasets for AI and robotic perception systems. My expertise includes bounding box annotation in YOLO format, dataset cleaning, and quality control, with a focus on electric vehicle charging connector detection projects. I am highly skilled in using tools such as LabelImg, Roboflow, CVAT, YOLOv5, and YOLOv8, and I am comfortable organizing large datasets for training, validation, and testing. With a strong engineering background and attention to detail, I ensure that the datasets I deliver are accurate, well-organized, and optimized for model performance. I am available for remote and freelance annotation projects and am committed to supporting long-term AI training data initiatives.

Labeling Experience

EV charging port classifications

RoboflowImageBounding BoxPolygon

This project focuses on developing an automated electric vehicle charging system using a robotic manipulator integrated with multi-sensor fusion, combining force sensing and computer vision. The computer vision system utilizes YOLOv8 to detect and classify CCS Type 2 charging ports and connectors in real time. Visual information from the camera is fused with force sensor feedback to guide the robot during the plug-in and unplug-in process, enabling precise alignment and safe interaction with the charging interface. The annotated image dataset plays a critical role in the system’s performance. High-quality labeled images of charging ports and connectors are used to train the YOLOv8 model, ensuring robust detection under varying distances, angles, and lighting conditions. The vision system allows the robot to localize the target accurately, while the force sensor provides real-time feedback to correct misalignment and prevent excessive contact force. This sensor fusion approach improv

This project focuses on developing an automated electric vehicle charging system using a robotic manipulator integrated with multi-sensor fusion, combining force sensing and computer vision. The computer vision system utilizes YOLOv8 to detect and classify CCS Type 2 charging ports and connectors in real time. Visual information from the camera is fused with force sensor feedback to guide the robot during the plug-in and unplug-in process, enabling precise alignment and safe interaction with the charging interface. The annotated image dataset plays a critical role in the system’s performance. High-quality labeled images of charging ports and connectors are used to train the YOLOv8 model, ensuring robust detection under varying distances, angles, and lighting conditions. The vision system allows the robot to localize the target accurately, while the force sensor provides real-time feedback to correct misalignment and prevent excessive contact force. This sensor fusion approach improv

2025 - 2025

Education

Work History