GitOps at Scale: What Breaks When You Hit 200 Repos and 40 Teams

Scaling GitOps from a single repo to 200 repos with 40 teams causes control plane meltdown, slow syncs, and drift. The ArgoCD controller gets overloaded, and manual kubectl fixes undermine trust in Git as source of truth. Solutions include sharding controllers, using webhooks, and raising reconciliation intervals for stable apps.