Conversational AI Tester / Dialogue Specialist
As a Conversational AI Tester and Dialogue Specialist, I evaluated large language model (LLM) conversations for instruction-following, emotional intelligence, and consistency. I adopted diverse user personas and stress-tested LLMs through multi-turn dialogues to identify behavioral failure modes. My responsibilities included thoroughly documenting model responses and generating detailed reports aligned with evaluation rubrics. • Designed and executed multi-turn conversations to test LLM instruction-following and emotion reasoning • Used adversarial and persona-based approaches to expose model weaknesses and edge cases • Identified, recorded, and tracked LLM behavioral patterns, failures, and inconsistencies • Maintained adherence to strict guidelines and structured reporting frameworks.