Freelance Virtual Assistant (AI output evaluation, prompt evaluation, RLHF-style feedback)
Provided structured evaluation and feedback on AI-generated text for quality, correctness, tone, and coherence. Performed RLHF-style preference ranking and prompt-response assessment to improve conversational AI outputs. Ensured responses met clarity and helpfulness expectations for downstream use. • Tone, accuracy, and coherence scoring with written rationale • Prompt-response rating for quality calibration • Preference ranking consistent with RLHF workflows • Feedback used to refine LLM behavior and conversational quality