Voudine O.

Key Skills

Software

Clickworker

Remotasks

Top Subject Matter

Scientific and clinical domains (STEM, biology, medicine, health sciences)

LLM response evaluation for STEM/biomedical content

Top Data Types

Text

Document

Image

Top Task Types

RLHF

Fine-tuning

Text Summarization

Evaluation/Rating

Prompt + Response Writing (SFT)

Question Answering

Freelancer Overview

AI Trainer, US-Based AI Training Technology Company. Brings 5+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Don't disclose, Internal, and Proprietary Tooling. Education includes Bachelor of Science, University of the People (2023) and Bachelor of Science, Kisii University (2025). AI-training focus includes data types such as Text and labeling workflows including RLHF, Evaluation, and Rating.

Labeling Experience

LLM Response Evaluator within AI Training role

Text

Performed response evaluation and ranking for LLM outputs to assess factuality, coherence, and safety compliance. Ensured that generated content met predefined standards before inclusion in training or QA cycles. Supported model optimization by repeatedly validating outputs against rubric-like constraints. • Reviewed LLM responses for factual accuracy and logical consistency. • Checked outputs for safety constraint adherence. • Reused and improved gold-standard guidance for better consistency. • Coordinated iterative feedback loops with cross-functional teams.

2024 - Present

AI Trainer, US-Based AI Training Technology Company

Don't discloseTextRLHF

Evaluated, ranked, and refined LLM-generated responses to ensure factual accuracy, logical coherence, and adherence to strict safety constraints. Generated high-quality, domain-specific training prompts and gold-standard responses across STEM topics, with emphasis on biology, medicine, and health sciences. Applied RLHF methodologies to optimize model behavior and reduce hallucination rates. • Conducted quality assurance reviews on peer annotations to maintain high data quality scores. • Collaborated with engineers and linguists to identify edge cases and debug model logic. • Improved NLP performance through iterative testing and refinement of training materials. • Produced training datasets and prompt-response pairs for downstream model evaluation.

2024 - Present

Education

Work History