Enhance LLM Response with RAG and Pinecone Integration