Vincent Kirui

Key Skills

Software

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

Text

Top Label Types

Classification

RLHF

Evaluation Rating

Prompt Response Writing SFT

Freelancer Overview

I am an experienced AI data solutions specialist with a strong background in academic content evaluation, data annotation, and training data creation for language models. My work includes producing high-quality, research-grade datasets focused on reasoning and analytical tasks in domains such as psychology, literature, and social sciences. I have evaluated and labeled long-form academic responses, designed prompts for humanities and social science research, and maintained consistent quality metrics above 98%. My technical skills include qualitative coding, SPSS, Python for data analysis, and the use of AI research assistants. I am adept at ensuring data integrity, scholarly rigor, and ethical standards throughout the data labeling process, and I thrive in multidisciplinary, detail-oriented environments.

Labeling Experience

LLM Text Annotation & Response Evaluation Project

Scale AITextClassificationRLHF

Worked on large-scale text-annotation pipelines for training and improving conversational AI systems. Tasks included categorizing model outputs by correctness, reasoning quality, tone, and safety; ranking multiple responses under RLHF workflows; and writing improved prompt-response pairs for supervised fine-tuning. Conducted secondary QA reviews on peer submissions, flagged hallucinations and guideline violations, and documented edge cases for project leads. Maintained high agreement rates with gold-standard samples and consistently met accuracy and throughput benchmarks.

Worked on large-scale text-annotation pipelines for training and improving conversational AI systems. Tasks included categorizing model outputs by correctness, reasoning quality, tone, and safety; ranking multiple responses under RLHF workflows; and writing improved prompt-response pairs for supervised fine-tuning. Conducted secondary QA reviews on peer submissions, flagged hallucinations and guideline violations, and documented edge cases for project leads. Maintained high agreement rates with gold-standard samples and consistently met accuracy and throughput benchmarks.

2023 - 2024

Education

Work History