AI Trainer & Data Annotator (Python-Focused, Remote)
I contributed to the development and curation of high-accuracy labeled datasets for NLP and Python code tasks as part of multiple remote AI projects. This involved crafting prompt-response pairs, evaluating LLM outputs for correctness and coherence, and annotating datasets for supervised machine learning pipelines. I ensured stringent quality assurance while handling large volumes of data and produced detailed feedback for fine-tuning model performance. • Regularly labeled data involving text, Python code, and multimodal datasets for technical domains. • Performed sentiment labeling, named entity recognition, intent classification, and text summarization review. • Evaluated model outputs in RLHF pipelines for factual accuracy, harmful content, and alignment with instructions. • Generated technically grounded and factually accurate training data through deep research.