Key Skills

Software

No software listed

Top Subject Matter

No subject matter listed

Top Data Types

Document

Image

Text

Top Label Types

Classification

Evaluation Rating

Red Teaming

RLHF

Text Generation

Freelancer Overview

While my professional background is rooted in Data Entry and evalution, I have spent the last several months pivot-training into AI Safety and Technical Evaluation. I specialize in identifying the 'Truth-Gap'—the space where an LLM sounds confident but is factually or logically incorrect. My approach is defined by a rigorous cross-referencing process; for example, when auditing legal-technical prompts, I don't rely on the model's fluency. Instead, I break down the reasoning steps to ensure the output aligns with primary sources and regulatory standards like the Colorado AI Act. What sets me apart is my focus on Edge-Case Detection. I am not just a labeler; I am a stress-tester. I have developed a personal workflow for Red-Teaming that involves pushing models to their logical limits to find 'hallucination drift' in professional prose. This disciplined attention to detail—ensuring that a 99% correct answer isn't disqualified by a 1% fatal error—is the standard I bring to every annotation and preference-ranking task.

Labeling Experience

Education

Work History