As healthcare systems move toward real-time data-driven operations, traditional diagnosis coding manual, retrospective, and delayed has become a bottleneck. Unlocking timely clinical insights requires new solutions that are both scalable and secure. To address this, we will explore the use of generative AI to automate diagnosis classification directly from clinical notes. Because fine-tuning open-source gen AI model requires large amount of high-quality training data, we will leverage GEMINI data to obtain real-world data distribution and feed these into proprietary large language model (LLM) to generate realistic synthetic notes. To ensure quality of the synthetic notes, combination of additional LLMs verification and human inspection will be performed. This approach offers a scalable path toward real-time diagnosis extraction, positioning LLMs as a strategic enabler for next-generation healthcare analytics.