Yangrui C.

Key Skills

Software

Labelbox

Label Studio

Top Subject Matter

Technology – LLM Training, Dialogue Systems & AI Safety

Logic & STEM – Chain-of-Thought Reasoning & Problem Solving

Creative Content – Natural Language Generation & Editing

Top Data Types

Text

Computer Code Programming

Top Task Types

RLHF

Freelancer Overview

Despite having just under a year of experience in AI training, I have developed a deep proficiency in Supervised Fine-Tuning (SFT) and RLHF (Reinforcement Learning from Human Feedback) workflows. I specialize in crafting high-complexity prompts and providing granular feedback on model outputs, focusing on accuracy, safety, and logical consistency. I am particularly adept at identifying "hallucinations" and refining model responses through Chain-of-Thought (CoT) reasoning, ensuring that the training data I produce directly contributes to more reliable and context-aware AI behavior.

Labeling Experience

AI Trainer & Data Quality Specialist

TextRLHF

I contributed to the optimization of a Large Language Model (LLM) through high-quality Supervised Fine-Tuning (SFT) and RLHF workflows. My work involved generating complex instruction-following datasets and evaluating multi-turn dialogues for accuracy, safety, and helpfulness. I specialized in Chain-of-Thought (CoT) reasoning tasks to improve the model's logical consistency in technical domains. During this period, I maintained a consistent 95%+ quality audit score, successfully identifying and correcting model hallucinations across thousands of prompts.

2025 - 2026

Education

Work History