Nathanael Thomas

Key Skills

Software

Data Annotation Tech

Top Subject Matter

AI-generated language

agentic workflows

LLM evaluation

Top Data Types

Text

Document

Image

Top Task Types

Evaluation/Rating

Classification

Text Generation

Question Answering

Text Summarization

RLHF

Fine-tuning

Red Teaming

Transcription

Data Collection

Function Calling

Prompt + Response Writing (SFT)

Freelancer Overview

Over 2 years freelance with DataAnnotation.tech working on the top-tier projects: Fact-checking long, information-dense responses, analyzing language and tone, evaluating and correcting agent workflows, creating entities and analyzing agent workflows within RL environments, reasoning about tool use and JSON schemas, comparing outputs across major LLMs, writing and correcting rubrics, providing structured feedback on complex multi-step tasks, and reviewing and editing submissions from other contributors.

Labeling Experience

AI Quality Specialist — Freelance

Data Annotation TechText

As an AI Quality Specialist, I conducted advanced evaluation work at the platform's highest tier. My responsibilities included evaluating, fact-checking, and correcting complex AI-generated responses and agentic workflows. This work required precision, critical thinking, and structured feedback across diverse evaluation and annotation tasks. • Fact-checked long, information-dense responses for accuracy and reasoning quality • Evaluated and corrected agent workflows, tool use implementations, and JSON schemas • Compared outputs across major large language models (LLMs) • Wrote rubrics and provided structured feedback on complex multi-step tasks

As an AI Quality Specialist, I conducted advanced evaluation work at the platform's highest tier. My responsibilities included evaluating, fact-checking, and correcting complex AI-generated responses and agentic workflows. This work required precision, critical thinking, and structured feedback across diverse evaluation and annotation tasks. • Fact-checked long, information-dense responses for accuracy and reasoning quality • Evaluated and corrected agent workflows, tool use implementations, and JSON schemas • Compared outputs across major large language models (LLMs) • Wrote rubrics and provided structured feedback on complex multi-step tasks

2024 - Present

Education

Work History