AI training & Hindi language quality evaluation (LLM response evaluation)
Evaluated Large Language Model (LLM) outputs for Hindi quality, accuracy, truthfulness, and safety compliance. Performed rigorous logical and detail-focused checks to ensure responses met expected linguistic and contextual standards. Applied prompt-engineering understanding to judge whether outputs aligned with the intended task and tone. • Assessed response correctness, clarity, and contextual nuance in Hindi • Verified factual consistency and adherence to safety/truthfulness guidelines • Reviewed large volumes of text with high visual attention to written detail • Provided structured feedback to improve model responses