I built a full-stack application where the frontend was generated and iterated using v0, while the entire backend logic and orchestration was implemented with n8n + AI matching
On the frontend side, v0 was used to rapidly design and refine the UI, focusing on user flows for authentication, property management, and data submission. The UI communicates exclusively with API endpoints, keeping the frontend lightweight and declarative.
The backend is fully handled by n8n, acting as an automation and integration layer instead of a traditional server. All API endpoints (login, user creation, listings, search, updates) are implemented as n8n webhooks, connected to modular workflows that handle validation, normalization, and persistence. Ask me for all workflows if you want to see them.
For data and identity management, the system integrates:
-
Firebase Authentication for secure user signup and login
-
Cloud Firestore as the primary database for users, listings, and relationships
Authentication tokens are validated server-side via n8n, while Firestore is used to store normalized domain data independently from auth records. This separation allows flexible role management (tenant, landlord, admin) and scalable workflows.
AI matching
The result is a serverless-style architecture where:
-
UI is fast to iterate (v0)
-
Business logic is visual, auditable, and extensible (n8n)
-
Auth and persistence are handled by Firebase
-
New features are added by extending workflows, not rewriting backend code