Building a Multi-Agent Valet Trash Command Center: Supabase Edge Functions, VAPI, and PostgreSQL at Scale

Quiet Hours Valet built a production command center that orchestrates eight concurrent AI voice agents and eight automation scenarios using Supabase Edge Functions, VAPI, and a self-hosted PostgreSQL stack. The system pairs a five-page Next.js frontend with an eight-table Postgres schema for scheduling and routing, and uses Supabase Edge Functions to handle VAPI webhook callbacks so a single operator can manage multiple autonomous agents at once. The build is presented as a technical reference for scaling voice AI beyond proof-of-concept, covering webhook-triggered database mutations and state management without Lambda overhead.

OPENING PARAGRAPH: We built a production command center for Quiet Hours Valet that orchestrates 8 concurrent AI voice agents, 8 automation scenarios, and real-time customer management across a vertically integrated real estate and services empire—all on Supabase, VAPI, and self-hosted infrastructure. This article breaks down the architecture: a 5-page Next.js frontend, 8-table Postgres schema for scheduling and routing, Supabase Edge Functions handling VAPI webhook callbacks, and the orchestration logic that lets a single operator manage multiple autonomous agents simultaneously. If you're scaling voice AI beyond proof-of-concept and need webhooks to reliably trigger database mutations, state management without Lambda overhead, and a schema that handles concurrent agent requests, this build is a technical reference.