Language Data Annotator
Contributed to large-scale AI and LLM training projects involving high-volume text and image data annotation for conversational AI systems. Responsibilities included intent classification, named entity recognition (NER), sentiment and emotion labeling, relationship classification, and evaluation of model-generated responses. Work also covered prompt–response writing (SFT), RLHF-style preference ranking, and quality evaluation to support model fine-tuning and performance improvement. Projects operated under strict annotation guidelines and confidentiality requirements. Quality measures included multi-pass review, ground truth verification, edge-case identification, and consistency checks to ensure high accuracy and reliability of training data. Delivered work within daily productivity targets while maintaining precision across multilingual datasets.