AI data annotation (prompt writing and AI response evaluation)
Created diverse multilingual prompts in Hindi (Devanagari and Roman Hindi) based on detailed annotation guidelines for AI behavior. Evaluated AI-generated responses by checking quality, relevance, factual accuracy, and compliance with the provided instructions. Applied analytical reasoning to language and STEM-related tasks while maintaining consistent labeling standards across assignments. • Multilingual prompt creation and instruction-following checks • Quality and relevance scoring / evaluation of model outputs • Factuality and adherence verification against guidelines • Consistency and attention-to-detail across annotation batches