Basic RAG Workflow

The Basic RAG Workflow template empowers you to build a simple AI-enhanced response system. This template leverages Scout’s hosted vector database and semantic search capabilities to extract relevant context from your content collections, then passes both the user query and the retrieved context to an LLM for a well-informed final answer.

Key features of this AI-powered template include:

• User Query Processing: Interprets natural language inputs to capture user intent accurately.
• Semantic Search Integration: Runs a semantic search on your Scout-hosted vector database to pull contextually relevant information.
• Contextual Output Generation: Combines the user's query with the retrieved context, allowing the LLM to generate coherent and insightful responses.
• Source Integration: Displays relevant links and citations from the retrieved content, ensuring transparency and easy reference.

This basic RAG workflow template is ideal for builders looking to get started with a basic RAG workflow that uses an LLM and a vector database. It handles a wide range of queries—from simple inquiries to complex issues—by efficiently merging direct user input with the deep contextual insights found in your Scout hosted vector database, which we call a Collection.

By adopting this Basic RAG workflow, you can easily get started building a RAG app without the technical headache or complex deployments. The template is fully customizable, enabling you to tailor the workflow to your specific brand voice and knowledge base.