Practical Tutorial Overview: Building an Intelligent Knowledge Retrieval and Q&A Experience
This tutorial focuses on the Knowledge module and RAG Pipeline capabilities in the AI Central NEXT platform, helping users master how to use the platform to build high-quality, orchestratable, and optimizable knowledge Q&A workflows through hands-on cases.
🧠 Case 1: Building a Knowledge Agent from Scratch
This practical case uses an AI technology company focused on large language model R&D as an example to demonstrate how to build an enterprise-grade intelligent Q&A system (Knowledge Agent) from scratch.
Through this tutorial, you will master the following key steps:
-
Create enterprise spaces and personal spaces
-
Build a dedicated knowledge base
-
Upload and process document content
-
Create an Agent and configure knowledge base connections
After completing the above process, you will have implemented an intelligent assistant with professional Q&A capabilities, helping enterprises efficiently activate the value of knowledge and drive internal collaboration and intelligent decision-making.
📚 Case 2: Intelligent Q&A for Multilingual Product Manuals
This practical case is designed to help users build an intelligent Q&A system that supports both Chinese and English product manuals. By leveraging the coordinated capabilities of preprocessing, retrieval, and plugins in the RAG Pipeline, it enables precise recall and high-quality answers for documents in different languages.
Through this tutorial, you will master the following key operations:
-
Upload multilingual product manuals and configure differentiated preprocessing Pipelines
-
Create a retrieval Pipeline that combines hybrid retrieval + language filtering + reranking
-
Use the Glossary terminology plugin to achieve terminology standardization
-
Configure the Agent and verify multilingual Q&A performance
This case is suitable for scenarios such as product documentation management, customer support, and technical support, helping enterprises provide consistent and accurate knowledge Q&A services for global teams.
Through learning this module, users can not only master the core configuration methods of the RAG Pipeline, but also understand the optimization path of "preprocessing quality → retrieval strategy → plugin extension", laying a solid foundation for building enterprise-grade knowledge services in the future.