Knowledge Base Management
Learn how to build and manage your AI assistant's knowledge base for accurate, relevant responses.
The Sensei AI Assistant uses a Retrieval Augmented Generation (RAG) system to provide accurate, contextual responses based on your content. Here's the process:
- 1Content Ingestion
Upload documents or crawl websites to add content
- 2Processing & Chunking
Content is split into optimized chunks for better retrieval
- 3Embedding Generation
Each chunk is converted to vector embeddings using OpenAI
- 4Vector Search
User queries are matched against embeddings to find relevant content
Content Sources
Automatically crawl and index entire websites or specific pages.
How to use:
- 1. Go to Admin → Knowledge Base
- 2. Click "Add Source" → "Website"
- 3. Enter the URL to crawl
- 4. Set crawl depth (number of pages)
- 5. Click "Start Crawling"
Upload PDF documents, text files, and other documents directly.
Supported formats:
- PDF documents
- Text files (.txt, .md)
- Word documents (.docx)
- CSV files
Maximum file size: 10MB per document
Do's
- • Keep content up-to-date and accurate
- • Use clear, concise language
- • Include FAQs and common questions
- • Organize content logically
- • Add product specifications and details
- • Include contact information where relevant
Don'ts
- • Don't include sensitive or private data
- • Avoid contradictory information
- • Don't use overly technical jargon
- • Avoid outdated content
- • Don't include personal customer data
- • Avoid copyrighted material
Keep your knowledge base current by regularly updating and reindexing your content.
When to reindex:
- • After major website updates
- • When product information changes
- • After adding new services or features
- • Quarterly for general maintenance
How to reindex:
- 1. Go to the source in your knowledge base
- 2. Click the "Reindex" button
- 3. Wait for the process to complete
- 4. Test with sample queries
Tips for better AI responses:
- →Use descriptive titles: Clear page titles and headings help the AI understand context
- →Include examples: Real-world examples help the AI provide more accurate responses
- →Be specific: Detailed product descriptions and specifications improve accuracy
- →Regular updates: Keep content fresh to maintain relevance