Coding Expertise Data labeler
Project Scope & Data Labeling Tasks: The project focused on evaluating the accuracy, coherence, and adherence of AI-generated responses from a Large Language Model (LLM). My role involved assessing responses based on predefined guidelines, including instruction following, localization, truthfulness, verbosity, and clarity. Project Size: The project spanned thousands of AI-generated responses, requiring meticulous review and structured feedback. I collaborated with a team of evaluators to ensure consistency across diverse language tasks and domains. Quality Measures Adhered To: Instruction Following: Ensuring AI responses align with explicit and implicit user instructions. Localization: Verifying cultural and linguistic appropriateness. Truthfulness: Fact-checking and identifying hallucinations or misinformation. Verbosity & Clarity: Balancing response length with readability and informativeness.