AI Customer Support Agent Evaluator (RAG System)
Text
Architected and evaluated a Retrieval-Augmented Generation (RAG) AI Customer Support Agent to improve knowledge retrieval and triage customer support tickets. Conducted sentiment analysis and assessed LLM outputs using structured evaluation criteria to enhance model responses. Maintained rigorous documentation for LLM performance, AI workflow integration, and user feedback analysis.
• Conducted iterative evaluations of AI chatbot outputs for accuracy and relevance
• Labeled user/customer support ticket sentiments and mapped responses to pre-defined categories
• Performed regular qualitative assessments of generative AI behavior for improvement cycles
• Applied best practices in model evaluation to support continual AI training and tuning.
Architected and evaluated a Retrieval-Augmented Generation (RAG) AI Customer Support Agent to improve knowledge retrieval and triage customer support tickets. Conducted sentiment analysis and assessed LLM outputs using structured evaluation criteria to enhance model responses. Maintained rigorous documentation for LLM performance, AI workflow integration, and user feedback analysis. • Conducted iterative evaluations of AI chatbot outputs for accuracy and relevance • Labeled user/customer support ticket sentiments and mapped responses to pre-defined categories • Performed regular qualitative assessments of generative AI behavior for improvement cycles • Applied best practices in model evaluation to support continual AI training and tuning.
2025 - Present