Infineon has identified the need for rapid access to product information for engineers, account managers, and customers. This problem is traditionally addressed using retrieval-augmented generation (RAG) chatbots. In this study, the author evaluates the RAG-Fusion method, which combines RAG with reciprocal rank fusion (RRF) by generating multiple queries, reranking them with reciprocal scores, and fusing the documents and scores. Through manual evaluation, the author found that RAG-Fusion provides accurate and comprehensive answers due to the generated queries contextualizing the original query from various perspectives. However, some answers strayed off-topic when the generated queries were insufficiently relevant to the original query. The research demonstrates significant progress in AI and NLP applications, particularly in providing technical information to engineers, sales-oriented information to account managers, and customer-facing information. The chatbot's performance was tested in three areas: answering product-specific questions from engineers, sales strategies from account managers, and customer queries about products. The chatbot excelled in accuracy and comprehensiveness but struggled with relevance and negative answers. Challenges include longer runtime due to the complexity of the LLM calls, answers straying from the original query, and the need for prompt engineering. Future work will focus on improving real-time performance, automated quality assurance, and multilingual support, particularly in non-English speaking countries. The author also plans to explore ways to better represent multimodal PDF datasheets as text for RAG-Fusion.Infineon has identified the need for rapid access to product information for engineers, account managers, and customers. This problem is traditionally addressed using retrieval-augmented generation (RAG) chatbots. In this study, the author evaluates the RAG-Fusion method, which combines RAG with reciprocal rank fusion (RRF) by generating multiple queries, reranking them with reciprocal scores, and fusing the documents and scores. Through manual evaluation, the author found that RAG-Fusion provides accurate and comprehensive answers due to the generated queries contextualizing the original query from various perspectives. However, some answers strayed off-topic when the generated queries were insufficiently relevant to the original query. The research demonstrates significant progress in AI and NLP applications, particularly in providing technical information to engineers, sales-oriented information to account managers, and customer-facing information. The chatbot's performance was tested in three areas: answering product-specific questions from engineers, sales strategies from account managers, and customer queries about products. The chatbot excelled in accuracy and comprehensiveness but struggled with relevance and negative answers. Challenges include longer runtime due to the complexity of the LLM calls, answers straying from the original query, and the need for prompt engineering. Future work will focus on improving real-time performance, automated quality assurance, and multilingual support, particularly in non-English speaking countries. The author also plans to explore ways to better represent multimodal PDF datasheets as text for RAG-Fusion.