Ezekiel Pascal

Key Skills

Software

Labelbox

Top Subject Matter

No subject matter listed

Top Data Types

Image

Video

Audio

Top Label Types

Bounding Box

Point Key Point

Segmentation

Classification

Tracking

Emotion Recognition

Freelancer Overview

I am an experienced AI Training Specialist with over three years dedicated to data labeling, annotation, and preparing high-quality training datasets for machine learning models in computer vision and natural language processing domains. My expertise includes using industry-leading tools like Labelbox, YOLO, DeepSORT, and CVAT to annotate images, videos, and audio for tasks such as object detection, tracking, and speech recognition. I have successfully led projects optimizing video datasets for real-time surveillance, annotated audio for speech-to-text models, and contributed to AI-powered traffic monitoring systems. I am skilled at collaborating with data science teams to ensure annotation accuracy and model readiness, and I continually strive to improve workflows and mentor others on best practices for scalable, reliable AI data solutions.

Labeling Experience

High-precision Multimodal Data Annotation for Computer Vision Models

LabelboxVideoBounding BoxPoint Key Point

Worked on large-scale video annotation projects for computer vision applications, focusing on object detection, tracking, and classification. Annotated frame-by-frame video sequences with bounding boxes, polygons, and segmentation masks, ensuring consistent ID assignment for multi-object tracking across all frames. Prepared datasets in YOLO and COCO formats for seamless integration into machine learning training pipelines. Applied detailed annotation guidelines and structured taxonomies to maintain consistency and accuracy across high-volume video data. Performed quality control through peer reviews and annotation audits, corrected labeling errors, and collaborated with ML engineers to refine guidelines and resolve edge cases. Delivered high-quality, training-ready video datasets for real-world computer vision model deployment.

Worked on large-scale video annotation projects for computer vision applications, focusing on object detection, tracking, and classification. Annotated frame-by-frame video sequences with bounding boxes, polygons, and segmentation masks, ensuring consistent ID assignment for multi-object tracking across all frames. Prepared datasets in YOLO and COCO formats for seamless integration into machine learning training pipelines. Applied detailed annotation guidelines and structured taxonomies to maintain consistency and accuracy across high-volume video data. Performed quality control through peer reviews and annotation audits, corrected labeling errors, and collaborated with ML engineers to refine guidelines and resolve edge cases. Delivered high-quality, training-ready video datasets for real-world computer vision model deployment.

2024

High-precision Multimodal Data Annotation for Computer Vision Models

LabelboxAudioBounding BoxPoint Key Point

Worked on large-scale audio annotation projects for AI training, including speech recognition, classification, and sound event detection. Labeled audio clips with time-stamped transcriptions, speaker identification, and emotion or intent classification. Segmented audio files into precise intervals for model training and ensured consistency across multiple annotators. Prepared datasets in formats compatible with ASR (Automatic Speech Recognition) and classification pipelines. Applied structured annotation guidelines, performed quality checks, and collaborated with ML engineers to refine labeling instructions and correct edge cases. Delivered high-quality, training-ready audio datasets for speech AI, voice assistants, and acoustic event detection models.

Worked on large-scale audio annotation projects for AI training, including speech recognition, classification, and sound event detection. Labeled audio clips with time-stamped transcriptions, speaker identification, and emotion or intent classification. Segmented audio files into precise intervals for model training and ensured consistency across multiple annotators. Prepared datasets in formats compatible with ASR (Automatic Speech Recognition) and classification pipelines. Applied structured annotation guidelines, performed quality checks, and collaborated with ML engineers to refine labeling instructions and correct edge cases. Delivered high-quality, training-ready audio datasets for speech AI, voice assistants, and acoustic event detection models.

2023 - 2025

High-precision Multimodal Data Annotation for Computer Vision Models

LabelboxImageBounding BoxPoint Key Point

Annotated large-scale image and video datasets for computer vision models focused on object detection, classification, segmentation, and multi-object tracking. Performed accurate bounding box and polygon labeling, as well as frame-by-frame tracking with consistent ID assignment to support MOT systems. Prepared datasets in YOLO and COCO formats for seamless integration into model training pipelines. Applied structured taxonomies and followed detailed annotation guidelines to ensure consistency across high-volume datasets. Maintained strong quality control standards through peer reviews, annotation audits, and dataset validation processes. Collaborated with ML engineers to resolve edge cases, refine labeling instructions, and deliver training-ready data aligned with production-level deployment requirements.

Annotated large-scale image and video datasets for computer vision models focused on object detection, classification, segmentation, and multi-object tracking. Performed accurate bounding box and polygon labeling, as well as frame-by-frame tracking with consistent ID assignment to support MOT systems. Prepared datasets in YOLO and COCO formats for seamless integration into model training pipelines. Applied structured taxonomies and followed detailed annotation guidelines to ensure consistency across high-volume datasets. Maintained strong quality control standards through peer reviews, annotation audits, and dataset validation processes. Collaborated with ML engineers to resolve edge cases, refine labeling instructions, and deliver training-ready data aligned with production-level deployment requirements.

2022 - 2024

Education

Work History