AI Trainer & LLM Output Evaluator
Text
As an AI Trainer & LLM Output Evaluator, I created and refined high-quality prompts and evaluated AI-generated outputs to improve LLM quality and safety. My work included rigorous fact-checking, writing detailed feedback, and following complex annotation rubrics. I consistently applied structured methodologies to ensure accuracy, coherence, and safety in model responses.
• Designed adversarial multi-turn prompt scenarios to expose and document model weaknesses.
• Scored and ranked outputs for accuracy, coherence, style, and safety adherence.
• Verified sources and checked claims for possible hallucinations or misinformation.
• Produced clear written rationales and maintained high inter-annotator agreement scores.
As an AI Trainer & LLM Output Evaluator, I created and refined high-quality prompts and evaluated AI-generated outputs to improve LLM quality and safety. My work included rigorous fact-checking, writing detailed feedback, and following complex annotation rubrics. I consistently applied structured methodologies to ensure accuracy, coherence, and safety in model responses. • Designed adversarial multi-turn prompt scenarios to expose and document model weaknesses. • Scored and ranked outputs for accuracy, coherence, style, and safety adherence. • Verified sources and checked claims for possible hallucinations or misinformation. • Produced clear written rationales and maintained high inter-annotator agreement scores.
2023 - Present