Cooper W.

Key Skills

Software

Labelbox

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Computer Code Programming

Image

Top Task Types

No task types listed

Freelancer Overview

I am a full-stack engineer with hands-on experience in building and optimizing data-driven systems using Django REST Framework and React. My recent work focused on developing a horse show management platform, where I independently implemented scoring algorithms (MOS Scoring), real-time judging interfaces for iPad, and API integrations with multiple organizations such as USEF and AHA. Through these projects, I developed strong skills in data structuring, annotation consistency, and model-driven architecture, which align closely with high-quality AI training data workflows. I am also proficient in using AI-assisted productivity tools like Cursor and Claude Code to improve efficiency, testing, and code accuracy. My approach emphasizes precision, data integrity, and clear process documentation — all essential qualities for effective AI data labeling and evaluation work.

Labeling Experience

AI camera/Horse Data Labeling

LabelboxImageBounding BoxData Collection

Worked on an AI camera–based dataset for horse show analysis and behavior tracking. Responsible for labeling horse and rider positions, motion sequences, and event actions across thousands of competition images and videos. Tasks included creating bounding boxes and keypoints for riders and horses, classifying posture and movement phases, and verifying detection accuracy. Ensured data consistency and high-quality annotations through multi-pass validation and adherence to strict quality metrics (IoU > 0.9, labeling precision > 98%). Collaborated with model engineers to refine labeling schemas and improve model detection accuracy for equestrian performance analysis.

Worked on an AI camera–based dataset for horse show analysis and behavior tracking. Responsible for labeling horse and rider positions, motion sequences, and event actions across thousands of competition images and videos. Tasks included creating bounding boxes and keypoints for riders and horses, classifying posture and movement phases, and verifying detection accuracy. Ensured data consistency and high-quality annotations through multi-pass validation and adherence to strict quality metrics (IoU > 0.9, labeling precision > 98%). Collaborated with model engineers to refine labeling schemas and improve model detection accuracy for equestrian performance analysis.

2025 - 2025

Education

Work History