Customer Support AI Use at Monte Carlo

Monte Carlo utilizes two separate Large Language Models (RunLLM Agents) to assist our support team.

MC Support Assistant

This AI Agent is used to provide first line support to customers via Slack and is trained only on publicly available documentation.

MC Internal Support Agent

This is a completely independent version of the AI Agent that is only used internally by Monte Carlo employees to help debug issues. This internal agent is trained on public documentation plus internal knowledge, which includes information shared by customers through Zendesk tickets created via emails to [email protected].

The model does not access or learn from any data within the Monte Carlo platform itself. It only uses what a customer shares in Zendesk tickets (e.g., messages, questions, screenshots).

Customer data is never used to train models for other customers and is not shared outside Monte Carlo or used to train RunLLM more broadly.