Roxanne J.

Key Skills

Software

Other

Top Subject Matter

AI training dataset QA and NLP annotation (sentiment, intent, NER tagging)

Content evaluation and NLP dataset annotation for AI training/fine-tuning

Legal Services & Contract Review

Top Data Types

Text

Document

Top Task Types

Fine-tuning

RLHF

Freelancer Overview

AI Data Annotator (NLP Logix): dataset annotation QA, sentiment/intent/entity tagging, and RLHF support. Brings 4+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Internal, Proprietary Tooling, and Other. Education includes Bachelor of Science, Wharton School (2018) and High School Diploma, West Orange High School (2014). AI-training focus includes data types such as Text and labeling workflows including Fine-tuning and RLHF.

Labeling Experience

AI Data Annotator (NLP Logix): dataset annotation QA, sentiment/intent/entity tagging, and RLHF support

TextFine-tuning

Performed QA checks and validation on annotated AI training datasets to ensure data integrity and label accuracy. Identified errors, edge cases, and inconsistencies in training examples, then supported dataset quality improvements. Conducted sentiment analysis, intent classification, and entity tagging as part of supervised labeling and downstream model training workflows. • Checked annotated datasets for correctness • Assisted RLHF workflows by enabling review and feedback loops • Labeled/organized structured and unstructured data for model learning • Used spreadsheets and annotation tools to manage labeling tasks.

2023 - 2025

Freelance Writer (PeoplePerHour/Fiverr): AI content evaluation plus RLHF-style feedback and text dataset labeling

OtherTextRLHF

Evaluated and refined AI-generated text content for clarity, factual accuracy, and alignment with client requirements. Completed data labeling and annotation for supervised learning datasets by categorizing text based on sentiment, intent, topic relevance, and contextual accuracy. Provided structured human feedback for RLHF-style training by identifying hallucinations, unclear phrasing, and inconsistent reasoning. • Labeled training data for sentiment, intent, topic relevance, and context • Performed comparative quality assessments across multiple AI outputs • Delivered RLHF-style feedback to improve response quality • Supported content safety and compliance checks for model outputs

2019 - 2019

Education

Work History