Step 1 — Project Setup
Configure your project name, domain, and use case in the RAG Wizard.
The first step of the RAG Wizard establishes the foundation for your RAG system by defining the project identity, domain, and intended use case.
Project Settings
Project settings define the basic identity and purpose of your RAG system. These values help the platform optimize retrieval strategies and provide context for how your knowledge base will be used.
| Setting | What It Does |
|---|---|
| Project Name | Identifies your RAG project — must be unique within your account |
| Description | Brief summary of what your knowledge base will be used for |
| Domain | Selects the primary subject area for your content, which helps optimize retrieval |
| Use Case | Defines how you plan to use the RAG system (chatbot, search, summarization, etc.) |
Domain Options
The domain selection helps the system apply domain-specific optimizations for retrieval. Choosing a focused domain improves accuracy because the system can tune chunking strategies, metadata extraction, and search parameters to match the content type.
| Domain | Best For |
|---|---|
| Customer Support | FAQ, help docs, support tickets |
| Knowledge Management | Internal wikis, company policies |
| Document Analysis | Contract review, legal analysis |
| Research | Academic papers, research notes |
| Content Creation | Marketing copy, blog research |
| Education | Course materials, study guides |
| General | Mixed or multi-domain content |
Use Case Options
The use case determines how end-users will interact with your RAG system. Different use cases require different response formats — a chatbot needs conversational answers while document search returns ranked snippets. Selecting your use case helps configure the pipeline for the expected interaction pattern.
| Use Case | Description |
|---|---|
| QA Chatbot | Question-answering chatbot interface |
| Document Search | Search across document collections |
| Content Summarization | Generate summaries from documents |
| Data Extraction | Extract structured data from text |
| Research Assistant | Multi-document research and analysis |