---
title: "Deploy ai-front-desk-template"
description: "AI receptionist that answers, texts back, books and invoices"
category: "AI/ML"
url: https://railway.com/deploy/ai-front-desk-template
---

# Deploy ai-front-desk-template

AI receptionist that answers, texts back, books and invoices

**[Deploy ai-front-desk-template on Railway](https://railway.com/template/ai-front-desk-template)**

Machine-readable deploy manifest (JSON, validated by TemplateCI): https://railway.com/deploy/ai-front-desk-template/manifest.json

- **Creator:** LeadSite AI
- **Category:** AI/ML

## Template content

### app

- **Source:** https://github.com/LeadGenius1/ai-offices-os
- **Public domain:** Yes

### Redis https://cdn.sanity.io/images/sy1jschh/production/0ce0bfdcfbdbf69662b1116671f97c2dd788b655-157x157.svg

- **Image:** redis:8.2
- **Start command:** `/bin/sh -c "rm -rf $RAILWAY_VOLUME_MOUNT_PATH/lost+found/ && exec docker-entrypoint.sh redis-server --requirepass $REDIS_PASSWORD --save 60 1 --dir $RAILWAY_VOLUME_MOUNT_PATH"`

### relay

- **Source:** https://github.com/LeadGenius1/ai-offices-os
- **Public domain:** Yes

### Postgres https://devicons.railway.app/i/postgresql.svg

- **Image:** ghcr.io/railwayapp-templates/postgres-ssl:18

## Documentation

# Deploy and Host AI Front Desk (AI Offices OS) on Railway

AI Front Desk is a self-hosted AI receptionist and front office for service businesses. Ava answers the phone, texts back missed calls, books appointments, sends review requests, runs follow-ups, and invoices through Stripe. It is multi-tenant, so one deployment can serve many businesses, and every outbound channel is off until you turn it on.

## About Hosting AI Front Desk (AI Offices OS)

This template deploys four services: the `app` (API, owner dashboard, scheduler, and the SMS, call, follow-up, review, campaign, billing, and support desks), the `relay` (a Twilio ConversationRelay WebSocket bridge for live voice, built from the `voice-relay/` directory of the same repository), a Postgres database (Prisma migrations run automatically before each start), and Redis (BullMQ queues and rate limits). All internal secrets are generated at deploy time. The app boots with only an Anthropic key and keeps each desk dark with HTTP 503 until that desk's provider key is present, so nothing sends, calls, or charges by accident. Per-provider and per-tenant daily spend ceilings are enforced by the built-in model router.

## Common Use Cases

- A plumber, HVAC, electrical, or cleaning company that misses after-hours calls and wants them answered, texted back, and booked.
- An agency or operator running a front desk for several local businesses from one deployment, each as its own tenant.
- A business that wants review requests, follow-ups, and Stripe invoices sent from one place with consent gating and spend caps.

## Dependencies for AI Front Desk (AI Offices OS) Hosting

- Anthropic API key (required for Ava's answers): set `AIOFFICES_ANTHROPIC_KEY` on `app`.
- Twilio account for the phone line and SMS: set `AIOFFICES_TWILIO_SID` and `AIOFFICES_TWILIO_TOKEN` on `app`, and `AIOFFICES_TWILIO_TOKEN` on `relay` (the relay verifies Twilio's signature on every WebSocket upgrade).
- Optional: ElevenLabs (`ELEVENLABS_API_KEY`, `AVA_TTS_VOICE`) for Ava's voice, Cal.com (`CALCOM_API_KEY`, `CALCOM_WEBHOOK_SECRET`) for bookings, Resend (`RESEND_API_KEY`, `RESEND_FROM`) for email, Stripe (`STRIPE_SECRET_KEY`, `STRIPE_WEBHOOK_SECRET`) for invoices. Stripe accepts test keys only unless `STRIPE_LIVE_OK=1`.

### Deployment Dependencies

- Repository: https://github.com/LeadGenius1/ai-offices-os
- Twilio ConversationRelay: https://www.twilio.com/docs/voice/twiml/connect/conversationrelay
- Anthropic API: https://docs.anthropic.com
- Prisma migrations run via `npx prisma migrate deploy` before the app starts.

### Implementation Details

After the first deploy, open the `app` domain and create the owner account. Owners and admins must enroll a TOTP factor on first login. Point your Twilio number's voice webhook at the app and its ConversationRelay at the relay's `wss://` domain (already wired into `AVA_VOICE_RELAY_WSS`). Turn desks on one at a time with the feature switches, all default off:

```
SMS_DESK_ENABLED=1
CALL_DESK_ENABLED=1
CAMPAIGNS_SEND_ENABLED=1
REVIEWS_SEND_ENABLED=1
BILLING_REMINDERS_ENABLED=1
```

Daily model spend is capped by `ROUTER_DAILY_CEILING_ANTHROPIC_USD` (default 15), `ROUTER_DAILY_CEILING_OPENROUTER_USD` (default 10), and `ROUTER_DAILY_CEILING_TENANT_USD` (default 5). The router refuses calls past the ceiling instead of running up a bill.

## Why Deploy AI Front Desk (AI Offices OS) on Railway?

Railway is a singular platform to deploy your infrastructure stack. Railway will host your infrastructure so you don't have to deal with configuration, while allowing you to vertically and horizontally scale it.

By deploying AI Front Desk on Railway, you are one step closer to supporting a complete full-stack application with minimal burden. Host your servers, databases, AI agents, and more on Railway.


## Similar templates

- [Chat Chat](https://railway.com/deploy/-WWW5r) — Chat Chat, your own unified chat and search to AI platform.
- [stella](https://railway.com/deploy/stella) — Self-host stella with web, API, Postgres, Redis, and object storage.
- [Hermes Agent | OpenClaw Alternative with Dashboard](https://railway.com/deploy/hermes-agent-or-openclaw-alternative-wit) — Self-Hosted Hermes AI Agent for Telegram, Discord & Slack

Open this page in a browser: https://railway.com/deploy/ai-front-desk-template