Abiral Parajuli

Data Annotator @Sigma AI

London, GB

Spearheaded the Quality Assurance (QA) and validation of AI-generated content for a major MAG7 technology client, covering complex text, image, and video datasets.• Functioned as a key escalation point for ambiguity, leveraging critical judgment to resolve edge cases and finalise guidelines when information was incomplete or contradictory.• Executed advanced data enrichment techniques, including long-form audio annotation, speaker diarisation, and linguistic labelling of non-speech events, to build robust \"ground truth\" models.• Drove immediate process and tooling improvements by identifying and reporting systemic bugs; implemented independent solutions to enhance annotation workflow efficiency.• Curated high-quality, production-ready data by structuring all linguistic and multimodal output into JSON format, directly fuelling the development of core NLP and STT algorithms.

University of Westminster

Master of Business Administration - MBA

Sichuan University

Bachelor's degree