conversational AI evaluation and response ranking
Worked on conversational AI training tasks involving evaluation and comparison of AI-generated text responses. Responsibilities included assessing instruction adherence, relevance to user intent, factual accuracy, tone, and overall helpfulness. Performed response ranking and quality classification using provided rubrics, identifying issues such as hallucinations, ambiguity, bias, or incomplete answers. Delivered consistent, structured judgements while adhering to quality guidelines and maintaining attention to detail across tasks.