<![CDATA[In today's digital era, the need for fast, accurate, and responsive information systems is increasingly pressing, especially in the higher education sector where prospective students often face obstacles in directly obtaining relevant and reliable academic information. To address this challenge, this research focuses on the design and development of an academic service chatbot by implementing the Retrieval-Augmented Generation (RAG) method at Catur Insan Cendekia University. The RAG approach combines information retrieval capabilities from various documents (retrieval) with the generative intelligence of language models, thus being able to produce contextual, personalized, and data-driven answers. The chatbot system was developed using Python, LangChain, FAISS, and the GPT model as the core of natural language processing. Performance evaluation was conducted using the ROUGE metric, which showed quite good results with a ROUGE-1 value of 0.50 and a ROUGE-L of 0.48. These findings prove that the system is capable of providing relevant and high-quality responses in helping answer prospective students' academic questions. With these advantages, this chatbot is expected to be an innovative solution to improve the quality of academic information services at UCIC, as it can present data quickly, accurately, interactively, and automatically. Furthermore, the implementation of this artificial intelligence-based technology aligns with digital transformation efforts in higher education, supporting the efficiency of academic services and strengthening the institution's image as a modern campus that adapts to developments in information technology.]]>
