LLM Training Data Annotation & Quality Review
Performed high-quality text annotation and evaluation to support large language model (LLM) training and fine-tuning workflows. Responsibilities included Named Entity Recognition (NER), text classification, prompt-response evaluation, and reinforcement learning from human feedback (RLHF) rating tasks. Annotated and reviewed diverse datasets including technical, cybersecurity, and general-domain content. Maintained strict quality standards through guideline adherence, cross-check validation, and consistency reviews. Utilized Labelbox and custom Python scripts to streamline annotation workflows and improve efficiency. Contributed to dataset accuracy improvements by identifying edge cases, ambiguous samples, and labeling inconsistencies.