Raava / CrewOps

01 · Case study

A CRM built for agents first.

CrewOps is a CRM for roofing and storm-restoration contractors. It follows a job from the first lead through the insurance claim, the build and final payment.

In productioniOS in review
CrewOps: a contractor's AI assistant works the records over MCP with a per-person token, while background agents route leads, link email and read scopes, in shadow until they match people's decisions. A contractor's AI assistant MCP · 9 tools · token scoped to one seat read · add · move CrewOps records leads · claims · jobs · docs Route leadagent Link emailagent Read scopeagent in shadow until they match what people decided
The assistant and the background agents work the same records the crew uses.

The assistant works the CRM

We built CrewOps agent-first. It has its own MCP server, so a contractor's AI assistant can read the pipeline, add prospects and notes, and move deals, through the same records the crew uses.

Each person mints their own token in the app and can revoke it. The token carries that person's seat: an assistant sees money only if its owner can, and only leadership can move a stage.

Agents behind the app

Small agents make the judgment calls: which rep a new lead goes to, which job an email belongs to, which document a scope came from. They start in shadow. Their recommendations are logged and compared with what people decided before they are allowed to act, and the manual flow stays in place if they fail.

Email in three tiers

  1. 1Link it automatically when the match is certain.
  2. 2Let a reasoning agent triage it when it is not.
  3. 3Hand it to a person when the agent is unsure.

Where it is

CrewOps is in production with its first contractor. Email ingestion, follow-up reminders and a scope reader are built and rolling out. An iOS app is in review.

Building something agents should run?

Raava Solutions takes on Palantir Foundry and AI agent work for clients. Tell us about the operation.

contact@raavasolutions.com