Joshua Ugwuasogwa

Key Skills

Software

Appen

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

Text

Top Label Types

RLHF

Evaluation Rating

Freelancer Overview

I am a detail-oriented AI trainer and data annotator with hands-on experience in evaluating, labeling, and improving natural language outputs for machine learning applications. My background includes systematically testing and annotating large language models such as GPT-4, Claude, and Gemini, with a strong focus on conversational AI evaluation, RLHF (Reinforcement Learning from Human Feedback), and content moderation. I have worked extensively with AI annotation platforms like Appen and Scale AI, using industry-standard guidelines to ensure data quality and consistency. My skills span prompt engineering, sentiment and intent classification, entity recognition, and data quality assurance, supported by advanced English proficiency and a solid foundation in scientific research methodology. I am passionate about leveraging human feedback to enhance AI safety and accuracy, and thrive in remote, task-based environments where attention to detail and critical thinking are essential.

Labeling Experience

Independent AI Research & Evaluation Contributor

Scale AITextEvaluation Rating

Independently tested and evaluated large language models (LLMs) such as GPT-4, Claude, and Gemini across a range of task types, documenting errors and model behaviors in detail. Developed hands-on experience in prompt engineering, human feedback cycles, and annotation best practices for AI model improvement. Prepared for professional data annotation and RLHF workflows through platform self-training and ethics study. • Conducted systematic comparative assessments of model outputs for factuality and safety. • Documented hallucinations, refusals, and error patterns in LLM responses. • Self-trained on annotation guidelines used by major data labeling platforms. • Practiced applying industry-standard quality assurance and RLHF processes.

Independently tested and evaluated large language models (LLMs) such as GPT-4, Claude, and Gemini across a range of task types, documenting errors and model behaviors in detail. Developed hands-on experience in prompt engineering, human feedback cycles, and annotation best practices for AI model improvement. Prepared for professional data annotation and RLHF workflows through platform self-training and ethics study. • Conducted systematic comparative assessments of model outputs for factuality and safety. • Documented hallucinations, refusals, and error patterns in LLM responses. • Self-trained on annotation guidelines used by major data labeling platforms. • Practiced applying industry-standard quality assurance and RLHF processes.

2023

Freelance Academic Research Assistant & AI-Assisted Tutor

AppenTextRLHF

Reviewed and annotated AI-generated text outputs for quality, accuracy, tone, and bias, closely following RLHF processes used by top AI companies. Assisted undergraduate students with academic projects using AI tools, involving evaluation and annotation activities supportive of model improvement. Crafted structured written feedback and identified factual errors, logical inconsistencies, and language issues in AI responses as part of quality assurance tasks. • Ensured factual accuracy and logical coherence in annotated outputs. • Used tools like ChatGPT, Claude, and Gemini to evaluate and annotate AI behaviors. • Maintained systematic records of research findings and annotations. • Translated complex concepts for clarity and instructional purposes.

Reviewed and annotated AI-generated text outputs for quality, accuracy, tone, and bias, closely following RLHF processes used by top AI companies. Assisted undergraduate students with academic projects using AI tools, involving evaluation and annotation activities supportive of model improvement. Crafted structured written feedback and identified factual errors, logical inconsistencies, and language issues in AI responses as part of quality assurance tasks. • Ensured factual accuracy and logical coherence in annotated outputs. • Used tools like ChatGPT, Claude, and Gemini to evaluate and annotate AI behaviors. • Maintained systematic records of research findings and annotations. • Translated complex concepts for clarity and instructional purposes.

2023

Education

Work History