Moses Solomon

Key Skills

Software

Other

Roboflow

Top Subject Matter

No subject matter listed

Top Data Types

Video

Top Label Types

Classification

Freelancer Overview

I am a detail-oriented professional with over 5 years of experience in data-driven environments, specializing in data labeling, annotation, and AI training tasks. My background includes hands-on work with image tagging, metadata attribution, sentiment analysis, and text classification, as well as conducting quality assurance to ensure high data accuracy and alignment with project guidelines. I am skilled in leveraging data categorization software, content management systems, and analytics tools like Google Analytics and SEMrush. My expertise in digital analytics and pattern recognition allows me to manage large datasets efficiently and deliver high-fidelity training data for AI and machine learning projects. I am passionate about supporting innovative AI solutions through precise data annotation and quality-focused workflows. Video Content Annotation: Leveraged the Atlas platform to review and validate short video segments, providing clear and accurate descriptive text annotations for model training. High-Precision Labeling: Maintained a 99% accuracy rate in data labeling tasks, adhering to strict project-specific protocols and quality standards. Quality Assurance (QA) Audits: Performed rigorous quality checks on team-labeled datasets within Atlas to ensure consistency and prevent data bias. Collaborative AI Training: Partnered with AI development teams to refine labeling taxonomies, improving the efficiency of model training cycles by 15%. Atlas-Specific Projects Dynamic Video Scene Tagging Project Goal: Train a Computer Vision model to recognize and label specific object movements in high-traffic urban video feeds. Tasks: Used Atlas to create frame-by-frame bounding boxes and temporal tags for object tracking. Impact: Contributed to a dataset of 10,000+ validated segments used for real-time safety monitoring models. LLM Concept Annotation (NLP) Project Goal: Categorize large-scale text corpora with human-understandable concepts to improve Large Language Model (LLM) interpretability. Tasks: Classified sub-document text into domains such as technology, philosophy, and law using structured labeling guidelines. Impact: Supported the development of more transparent and auditable AI responses through fine-grained data control.

Labeling Experience

Image annotation expert

RoboflowImagePolygonSegmentation

Image & Video Annotation Experience ​2D Bounding Boxes & Polygons: ​Performed high-precision labeling of urban traffic scenes, identifying vehicles, pedestrians, and cyclists. ​Utilized polygon tools to trace irregular shapes like construction zones and road debris to assist in path-planning AI training. ​Semantic Segmentation: ​Conducted pixel-level labeling to distinguish between drivable surfaces (roads) and non-drivable surfaces (sidewalks, curbs, and vegetation). ​Ensured 100% boundary accuracy to prevent "noise" in the machine learning model. ​Object Tracking (Video): ​Handled temporal consistency tasks, ensuring that a single "Object ID" followed the same vehicle across multiple video frames, even during partial occlusions. ​Annotated frame-by-frame state changes, such as brake light activation and turn signal usage. SKILLS Annotation Types: Bounding Boxes, Polygons, Keypoint Labeling, Semantic Segmentation. Tools: CVAT, LabelImg, Labelbox, or proprietary internal tools. Attrib

Image & Video Annotation Experience ​2D Bounding Boxes & Polygons: ​Performed high-precision labeling of urban traffic scenes, identifying vehicles, pedestrians, and cyclists. ​Utilized polygon tools to trace irregular shapes like construction zones and road debris to assist in path-planning AI training. ​Semantic Segmentation: ​Conducted pixel-level labeling to distinguish between drivable surfaces (roads) and non-drivable surfaces (sidewalks, curbs, and vegetation). ​Ensured 100% boundary accuracy to prevent "noise" in the machine learning model. ​Object Tracking (Video): ​Handled temporal consistency tasks, ensuring that a single "Object ID" followed the same vehicle across multiple video frames, even during partial occlusions. ​Annotated frame-by-frame state changes, such as brake light activation and turn signal usage. SKILLS Annotation Types: Bounding Boxes, Polygons, Keypoint Labeling, Semantic Segmentation. Tools: CVAT, LabelImg, Labelbox, or proprietary internal tools. Attrib

2025

Data Labeling & Strategy Manager / Annotation & Data Specialist

OtherVideoClassification

Oversaw and executed data labeling and annotation processes to support AI model training and development. Ensured content categorization and sentiment analysis met quality standards and project objectives. Utilized digital tools and team leadership to optimize annotation workflows and maintain accuracy. • Conducted regular quality assurance checks on team-labeled data. • Managed large datasets for content tagging and classification. • Specialized in text classification and sentiment analysis tasks. • Leveraged CMS and data categorization software for streamlined annotation.

Oversaw and executed data labeling and annotation processes to support AI model training and development. Ensured content categorization and sentiment analysis met quality standards and project objectives. Utilized digital tools and team leadership to optimize annotation workflows and maintain accuracy. • Conducted regular quality assurance checks on team-labeled data. • Managed large datasets for content tagging and classification. • Specialized in text classification and sentiment analysis tasks. • Leveraged CMS and data categorization software for streamlined annotation.

2023

Education

Work History