Rentals · Customer supportLocService is a peer-to-peer property rental platform that handles several hundred customer-support tickets a day. We built an assistant that reads all their past tickets and internal documentation to draft the answer for each new ticket, tailored to the customer's profile and history. The AI prepares, a person sends: the support team stays in charge of every reply.
We did not build a chatbot that answers customers on its own. On a support desk where the right answer depends on each customer's file, an AI mistake shows right away, and it goes out with LocService's name on it. So the assistant does the research and the writing, and the team keeps the decision. What made the drafts right was not a clever prompt: it was the real customer context pulled from their systems, and rules that recognize the case before drafting.
Coup de Pates supplies frozen products to food professionals. We built them Ensemble, the app that brings employees together around events like a step challenge, and a library of AI skills that checks their product documents.
Read the case→
The Vibe Company supports Theodo as a technical expert on critical projects for their clients, such as building Software Factories for hundreds of developers, to help accelerate their teams.
Monka supports people who care for a loved one, alongside coordinating nurses. To keep up with their growth, we shipped in two months the caregiver app, Lifeline, the CRM for their nurses, and the API that ties them together.
To go further, read our guide AI agents for customer support: automate without hurting service.
The Vibe Company is an AI agency in Paris: AI products, agents and training, all the way to production.
LocService handles several hundred support tickets a day on its peer-to-peer rental platform. We built an assistant that drafts the answer for each new ticket.
It reads past tickets and internal docs, connects to internal APIs for real customer context, applies classification rules, then drafts a reply tailored to the customer’s profile and history.
A person. A support agent reviews the draft, fixes it if needed, then sends it. The AI prepares; the team keeps the decision.
87% of tickets are answered with our AI drafts. The day-to-day measure: is the draft used?