What the assistant actually does.
Most WhatsApp AI products are sold as a self-serve platform: subscribe, connect a number, deploy an agent in minutes. Kavya is not that. It is the assistant we built and run ourselves, and what we describe below is how it actually works, then what a version scoped to your business would look like.
A real qualification conversation, not a menu
A prospect messages a WhatsApp number and Kavya opens by disclosing she's an AI system, in the language the prospect wrote in, not buried in a menu but in the first message. From there she asks the questions that would otherwise take a person fifteen minutes to establish, and hands the conversation to a human the moment it clears that bar.
One conversation, in whichever language a prospect opens with
Kavya supports nine languages, the same nine askquorumai.com is available in, detected from a prospect's opening message and never re-guessed mid-conversation. A visitor who opens in French stays in French for the rest of the exchange, including the handoff, because switching languages mid-conversation reads as a product that isn't actually multilingual.
Qualification and handoff, never a sales pitch
Kavya's job ends where a real conversation begins. She asks a closed set of qualification questions, never improvises pricing or capability claims, and says plainly when she doesn't know something rather than guessing. Every conversation that clears the qualification bar goes to a person. She doesn't close deals, and she isn't supposed to.
Built to disclose, by law and by default
Kavya discloses that she's an AI system at the start of every conversation, in the language of that conversation, which is what Article 50 of the EU AI Act requires, and what we'd do regardless of what the law says. It's designed into the same pass that generates the opening message, not a separate step that can silently be skipped.