AI trainer
Worked on an AI training and evaluation project focused on an insurance claims assistant chatbot. My role involved having natural conversations with the AI to test how accurately it handled real-world insurance tasks such as verifying policies, creating claims, assessing vehicle or property damage from images, checking claim status, and scheduling inspections. I also reviewed the model’s responses for conversational quality, safety compliance, and overall user experience, while documenting errors and providing ideal function-call workflows to help improve the system’s accuracy and reliability.