Large Language Models excel in answering medical exam questions, but applying them to real-world problems often requires more guidance and fine-tuning for precise results. In this lecture, I'll review existing methods for effectively applying LLMs to medical question answering tasks and demonstrate how to use Azure's Text Analytics for Health to improve their performance and effectively evaluate the results they generate.