Jenie Ranger-pepin

Key Skills

Software

Labelbox

Other

Internal/Proprietary Tooling

Top Subject Matter

No subject matter listed

Top Data Types

Document

Text

Top Task Types

Evaluation Rating

Fine Tuning

Prompt Response Writing SFT

RLHF

Text Generation

Freelancer Overview

I have worked as an AI Trainer for a tech company, contributing to the development of advanced language models through data labeling, prompt generation, and prompt comparison tasks. Fluent in French, English, and Italian, I played a key role in multilingual training processes, ensuring high-quality, context-aware datasets across languages. My work involved crafting and evaluating prompts that aligned with linguistic and cultural nuances, helping improve model performance and relevance in real-world applications. In addition to content creation, I specialized in evaluating outputs for factual inaccuracy, hallucination, and potentially harmful or dangerous content. My academic background in education, philosophy, psychology, and political science enabled me to assess complex information with critical insight and precision. This interdisciplinary expertise allowed me to contribute to safer, more accurate AI systems by identifying content risks and ensuring responses met ethical and factual standards.

Labeling Experience

AI Advanced trainer

LabelboxTextRLHFFine Tuning

Thank you! Based on your answers, here’s a well-formed response: --- The project focused on training large language models (LLMs) and chatbots within specialized domains such as education, the humanities, academic writing, and language translation. As part of this initiative, I performed a range of advanced data labeling tasks, including relevance ranking, evaluation of factual inaccuracy, hallucination detection, and the identification of potentially dangerous content. My work contributed to refining model outputs for high-stakes, knowledge-intensive areas where precision and ethical standards are critical. The dataset was extensive, and I collaborated with a large, multilingual team to ensure comprehensive coverage and consistency across tasks. The project maintained rigorous quality control processes, including regular reviews and performance tracking. My annotations consistently exceeded quality benchmarks and were frequently rated as flawless.

Thank you! Based on your answers, here’s a well-formed response: --- The project focused on training large language models (LLMs) and chatbots within specialized domains such as education, the humanities, academic writing, and language translation. As part of this initiative, I performed a range of advanced data labeling tasks, including relevance ranking, evaluation of factual inaccuracy, hallucination detection, and the identification of potentially dangerous content. My work contributed to refining model outputs for high-stakes, knowledge-intensive areas where precision and ethical standards are critical. The dataset was extensive, and I collaborated with a large, multilingual team to ensure comprehensive coverage and consistency across tasks. The project maintained rigorous quality control processes, including regular reviews and performance tracking. My annotations consistently exceeded quality benchmarks and were frequently rated as flawless.

2023

Education

Work History