for those who have been running large platforms cir + 100 dev teams, how do you document the platform architecture, how do you keep the documentation up to date when your infra is constantly changing, what tools are you using?

We’ve introduced RFC process at several organizations to keep evolution of the platform transparent to the users.

We use an RFC design/approval process, ADRs living with code (IaC and services) and lastly tools like Cloudcraft (AWS) or Clodockit (Multi-cloud) for live views of our infra. These require disciplined tagging to filter & visualize specific “stack” deployments. We are evaluating Firefly for cataloging and assessing drift/unmanaged/untagged resources.