RAG Knowledge Assistant

Ask questions about company policies and documentation. Answers include citations and confidence scores.

Chat

Strict Mode

Ask a question to get started

Sample Questions

Click to try these examples

How It Works

1. Retrieval: Your question is embedded and matched against document chunks using vector similarity.

2. Context: Top-k relevant chunks are passed to the LLM as context.

3. Generation: The LLM generates an answer based only on the provided context.

4. Citations: Source documents are included for verification.

Strict Mode: When enabled, low-confidence answers return "I don't know" instead of guessing.