AI Response Evaluator for Educational Platform
Evaluated AI-generated responses for a study assistant platform. Adjusted prompts to improve the relevance and accuracy of answers provided by the AI system. Worked with Canvas LMS and OpenAI API in a workflow involving retrieval-augmented generation and course-specific data. • Assessed the quality of generative text outputs for pedagogical relevance • Provided structured feedback to guide AI prompt engineering • Contributed to iterative improvements of an educational LLM assistant • Employed version-controlled workflows and tracked prompt modifications for performance gains