Agents in action
Rillet's two-person frontend team closed 800+ Linear tickets in 4 months with eve agents
Rillet, an AI-native ERP platform, built internal agents with Vercel's eve framework so its customer success team can request product changes that a coding agent turns into pull requests. A two-person frontend team reviews and merges them.
What the agent did
- A Chrome extension agent captures what a customer sees. If the change is safe and well-scoped, it passes the request to a coding agent.
- The coding agent writes the spec, makes the change and opens a pull request assigned to Anthony Liang, who reviews and merges it.
- Agents open a browser in a sandbox to test each change and attach screenshots as proof. If changes are requested, the agent resumes its session.
- Rillet's CPA team ships customer requests through its in-house agent Nebula within two hours of the ask.
- Rillet says its shipping rate tripled over three months.
Why it matters
Customer requests used to scatter across a Linear ticket, a Slack thread and a conversation with an engineer. Letting the person closest to the customer start the change, with one engineer reviewing, shortens that path. A shared agent framework also lets non-TypeScript staff shape agent behavior through markdown instructions.
This is a Vercel customer story, so the figures come from Vercel and Rillet, and an engineer reviews and merges every pull request.
Source: Vercel
Every number on this page comes from the source. Our charts redraw those numbers; videos are embedded from the publisher.
