Yamkela V.

Key Skills

Software

CVAT

Top Subject Matter

Computer Vision/Object Detection

Legal Services & Contract Review

Regulatory Compliance & Risk Analysis

Top Data Types

Image

Text

Document

Top Task Types

Bounding Box

Freelancer Overview

I’ve spent 3+ years working hands-on in data labeling and AI training data across NLP, computer vision, and multimodal projects. My core focus has been creating high-quality, consistent datasets for LLMs and classification models—handling tasks like intent classification, entity annotation, sentiment labeling, instruction-response ranking, and image segmentation. I’m experienced with annotation guidelines, QA/QC workflows, and inter-annotator agreement, so I catch edge cases and reduce noise before data hits training. What sets me apart is combining attention to detail with domain adaptability. I’ve worked on projects ranging from conversational AI for customer support to safety labeling for content moderation, and I pick up new schemas quickly. I’m comfortable using Labelbox, CVAT, and custom internal tools, and I document ambiguities clearly to improve guidelines. Beyond labeling, I bring strong communication and calibration skills—I’ve led small review sessions to align teams and reduce drift. I’m comfortable with high-volume targets but won’t sacrifice accuracy for speed, because I know bad data costs more later.

Labeling Experience

Visual Data Annotator

CVATImageBounding Box

Over four years, I labeled, categorized, and verified over 10,000 images and video frames for AI and machine learning computer vision projects. My work focused on bounding box and polygon annotation, object detection, and semantic segmentation. I consistently maintained accuracy above 98% and upheld data privacy and compliance standards. • Performed quality assurance on annotated datasets to ensure high-quality deliverables. • Trained junior annotators on annotation tools and guidelines. • Collaborated closely with ML engineers to enhance data quality. • Used key annotation software including CVAT, Labelbox, VGG Image Annotator, and SuperAnnotate.

2020 - 2024

Education

Work History