Arnold Kanani

Key Skills

Software

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

Image

Top Label Types

Bounding Box

Freelancer Overview

I am a detail-oriented AI data labeling and training specialist with hands-on experience in annotating and preparing high-quality datasets for machine learning projects. My work spans text, image, and document labeling, including tasks such as text classification, sentiment tagging, and entity identification. I am skilled at following complex guidelines, ensuring data accuracy, and performing thorough quality assurance and validation. I have a strong background in data cleaning, content moderation, and handling confidential information, and I am proficient with tools like Microsoft Excel, Google Sheets, and CRM systems. Working remotely with global teams has strengthened my communication and collaboration skills, and I am committed to continuous learning and maintaining high standards of data integrity and accuracy.

Labeling Experience

Data annotator

Scale AIImageBounding Box

Project: LLM-Enhanced Retail Data Annotation Contributed to a project improving an LLM powered e-commerce platform by evaluating AI-generated product search responses and labeling retail data. Tasks included response ranking, hallucination detection, product categorization and attribute tagging (brand, color, size). The datasets covered 50,000 combined text and product samples over 10 weeks. Quality was ensured through detailed guidelines, double reviews (10-15),≥85% inter-annotator agreement, and ongoing QA audits.

Project: LLM-Enhanced Retail Data Annotation Contributed to a project improving an LLM powered e-commerce platform by evaluating AI-generated product search responses and labeling retail data. Tasks included response ranking, hallucination detection, product categorization and attribute tagging (brand, color, size). The datasets covered 50,000 combined text and product samples over 10 weeks. Quality was ensured through detailed guidelines, double reviews (10-15),≥85% inter-annotator agreement, and ongoing QA audits.

2025 - 2025

Education

Work History