Billy Ibrahim Adhipati Djayanegara

Key Skills

Software

Labelbox

Top Subject Matter

No subject matter listed

Top Data Types

Text

Video

Top Label Types

Data Collection

Text Generation

Translation Localization

Freelancer Overview

I am a results-oriented professional with a strong foundation in Government Science and a specialized focus on Human Resources. I have direct, hands-on experience in AI data generation through "Project Terminator," where I specialized in high-fidelity egocentric (POV) video collection for Computer Vision and Human Activity Recognition (HAR). This role required strict adherence to complex technical protocols, including specific camera angles (45-degree downward tilt), frame rate adjustments, and meticulous environmental staging to ensure data quality for model training. I am proficient in using labeling platforms such as Labelbox and am experienced in managing rework workflows to meet rigorous quality standards.

Labeling Experience

Project Terminator

LabelboxVideoClassificationObject Detection

Contributed to the development of a large-scale dataset for egocentric (first-person) domestic activity recognition. Responsibilities included performing classification and descriptive labeling of activities and recording high-quality videos of natural everyday chores using a head-strap mount. I ensured that all movements were smooth and that objects being manipulated (picking, holding, placing) remained fully in-frame. I strictly followed technical mandates for camera settings (30 FPS, 720p+ resolution) and privacy protocols to ensure no faces or sensitive information were captured. I also managed the rework queue in Labelbox to address quality feedback and resubmit corrected data rows.

Contributed to the development of a large-scale dataset for egocentric (first-person) domestic activity recognition. Responsibilities included performing classification and descriptive labeling of activities and recording high-quality videos of natural everyday chores using a head-strap mount. I ensured that all movements were smooth and that objects being manipulated (picking, holding, placing) remained fully in-frame. I strictly followed technical mandates for camera settings (30 FPS, 720p+ resolution) and privacy protocols to ensure no faces or sensitive information were captured. I also managed the rework queue in Labelbox to address quality feedback and resubmit corrected data rows.

2025

Education

Work History