AI Data Annotator & Trainer (LLM evaluation, prompt + response training support) — Freelance/Remote (RWS WorkZone)
Rated AI-generated responses to assess factual accuracy, logical consistency, and compliance with safety guidelines for LLM refinement. Authored, reviewed, and optimized complex prompts to support AI training in technical reasoning and domain-specific knowledge. Produced high-quality labeled/annotated dataset contributions aligned to project quality benchmarks and documentation expectations. • Evaluated response quality using rubrics for accuracy, coherence, and safety adherence • Drafted and optimized prompts for improved model reasoning and domain performance • Performed dataset labeling/annotation with quality compliance checks • Supported iterative training improvements through prompt/response feedback cycles