Core Features
Knowledge Base & Vector Search
Unlock accurate AI answers by feeding Sentrup your actual business files. The system indexes your documents into a secure vector space and matches them to incoming customer questions in real time.
Supported Document Formats
Sentrup processes text inputs from multiple sources:
- PDF Documents: Manuals, product specifications, and policy sheets.
- Text & Markdown files: TXT and MD files detailing product instructions or FAQs.
- URL Web Crawls: Provide page URLs, and Sentrup will automatically scrape and index their text content.
Chunking & Vector Embeddings
When you upload a file, the platform does not read it as one huge text chunk. The pipeline:
- Splits the text into overlapping chunks (typically 500-1000 characters) to preserve contextual boundaries.
- Generates vector embeddings using state-of-the-art embedding models, transforming text chunks into multi-dimensional numerical coordinate vectors.
- Saves coordinate vectors into a dedicated secure index in the Qdrant vector database.
Preventing AI Hallucinations
Unlike public chat engines that speculate when they do not know the answer, Sentrup strictly bounds response scopes.
When a visitor submits a question, Sentrup performs a cosine similarity calculation inside Qdrant to find the top 3-5 matching text chunks. These chunks are injected into the LLM context. If no matching chunks exceed your similarity confidence threshold, the system automatically responds with a customizable fallback message or escalates the conversation to a human support agent.