AI Engineer Intern — Kapital Bank
As an AI Engineer Intern at Kapital Bank, I evaluated outputs from large language models and conversational AI systems using defined quality rubrics. I assessed the accuracy, coherence, and completeness of model-generated responses across multiple digital channels. I reviewed agent workflows and NLU pipeline traces to identify errors and provided actionable feedback. • Evaluated LLM and conversational AI responses against rubrics • Identified systematic errors, hallucinations, and intent misclassifications • Documented edge cases and recurring failure modes for model refinement • Adapted evaluation to ambiguous guidelines and novel scenarios