How the AI Works

ResolveBot uses RAG (Retrieval-Augmented Generation) to answer questions accurately.

The Pipeline

1. Embedding — Customer message converted to a vector
2. Vector Search — Top 5 matching knowledge base passages retrieved
3. Response — GPT-4o mini generates an answer using only those passages
4. Confidence — Score calculated with heuristic adjustments
5. Action — Auto-send, draft, or escalate based on score

Cost Per Ticket

~$0.006 (less than 1 cent). Included in your plan — no per-ticket charges.

Language

The AI detects the customer's language and responds in the same language. Vector search works across languages.