AI Data Trainer / RLHF Evaluator (Mindrift)
Performed RLHF evaluation by reviewing and ranking AI-generated responses for quality, accuracy, and safety. Identified and flagged harmful, biased, or factually incorrect outputs according to project guidelines. • Evaluated responses across three tasks per week • Maintained a 95% quality score • Wrote high-quality prompts and ideal responses to support supervised fine-tuning • Collaborated asynchronously with QA teams to meet deadlines