Christiana Etuk

Key Skills

Software

Data Annotation Tech

Top Subject Matter

No subject matter listed

Top Data Types

Text

Top Label Types

Question Answering

RLHF

Evaluation Rating

Prompt Response Writing SFT

Freelancer Overview

I am an experienced AI Data Trainer and Content Reviewer with a strong background in data annotation, response evaluation, and content moderation. My work has included labeling text data, evaluating AI-generated responses for accuracy and compliance, and providing feedback to enhance model performance. I am skilled in using tools such as Google Workspace, Microsoft Office Suite, Slack, CRM systems, and task management platforms, and I consistently meet high quality and accuracy standards in remote environments. My attention to detail, fast typing speed, and ability to follow detailed guidelines have enabled me to maintain over 95% task accuracy and deliver reliable results in NLP and data labeling projects. I am also adept at reviewing peer submissions, flagging harmful content, and ensuring data integrity throughout the annotation process.

Labeling Experience

AI Response Evaluation and Rating Practice Project

Data Annotation TechTextQuestion AnsweringRLHF

Conducted structured AI response evaluation practice to simulate real world data labeling workflows for large language models. Generated diverse prompts across customer service, research, and general knowledge scenarios. Compared multiple AI generated responses and rated them based on accuracy, clarity, helpfulness, tone, and safety. Identified hallucinations and factual inconsistencies through independent web verification. Flagged biased or unsafe outputs using standard AI safety principles. Rewrote weak responses to improve logic, completeness, and user relevance. Documented structured justifications for every rating decision to mirror RLHF evaluation processes. Maintained internal accuracy tracking above 95 percent across over 150 evaluated prompt response pairs.

Conducted structured AI response evaluation practice to simulate real world data labeling workflows for large language models. Generated diverse prompts across customer service, research, and general knowledge scenarios. Compared multiple AI generated responses and rated them based on accuracy, clarity, helpfulness, tone, and safety. Identified hallucinations and factual inconsistencies through independent web verification. Flagged biased or unsafe outputs using standard AI safety principles. Rewrote weak responses to improve logic, completeness, and user relevance. Documented structured justifications for every rating decision to mirror RLHF evaluation processes. Maintained internal accuracy tracking above 95 percent across over 150 evaluated prompt response pairs.

2025

Education

Work History