---
title: "Deploy pilotdeck"
description: "Open-source AI Agent operating system for task-oriented productivity"
category: "AI/ML"
url: https://railway.com/deploy/pilotdeck
---

# Deploy pilotdeck

Open-source AI Agent operating system for task-oriented productivity

**[Deploy pilotdeck on Railway](https://railway.com/template/pilotdeck)**

- **Creator:** kmaki's Projects
- **Category:** AI/ML
- **Total deploys:** 1

## Template content

### pilotdeck

- **Image:** xiaosong233/pilotdeck-railway:latest
- **Public domain:** Yes

## Documentation

# Deploy and Host PilotDeck on Railway

## About Hosting PilotDeck on Railway

PilotDeck is an open-source AI Agent operating system that runs as a containerized application. Railway provides a simple, one-click deployment option for PilotDeck with persistent storage for agent workspaces, configuration, and memory. The service runs two cooperating Node.js processes: a gateway for agent runtime and a UI server for web access.

## Tech Stack

- **Node.js 22** - Runtime environment
- **TypeScript** - Language
- **Vite** - Frontend build tool
- **SQLite** - Embedded database for auth and sessions
- **PilotDeck Gateway** - Agent execution runtime
- **OpenRouter API** - AI model integration (configurable)

## Why Deploy PilotDeck on Railway

Railway offers several advantages for hosting PilotDeck:

- **One-click deployment** with no configuration required
- **Automatic scaling** based on resource needs
- **Persistent storage** for PilotDeck state, configuration, memory, and projects
- **Public domain** with automatic HTTPS
- **Environment variables** for easy model/API key configuration
- **Docker-native** deployment with zero vendor lock-in

## Common Use Cases

- **AI-assisted development** - Use agents to help with coding and project analysis
- **Task automation** - Build agents that handle repetitive workflows across multiple projects
- **Knowledge management** - Leverage PilotDeck's memory system to maintain context across tasks
- **Multi-agent orchestration** - Run multiple specialized agents in parallel on long-running work

## Deployment Notes

- **Port**: PilotDeck UI listens on port 3001 (configured via `SERVER_PORT` env var)
- **Gateway Port**: Internal gateway runs on 18789 (configured via `PILOTDECK_GATEWAY_PORT`)
- **Volume Persistence**: `/root/.pilotdeck` directory persists all state including config, auth DB, skills, plugins, and memory
- **Model Configuration**: Uses OpenRouter by default with Deepseek v4 Flash (configurable via `PILOTDECK_MODEL`, `PILOTDECK_API_KEY`, `PILOTDECK_API_URL`)
- **Light Model**: Qwen 3 8B for token-saving tier routing (configurable via `PILOTDECK_LIGHT_MODEL`)
- **Health Check**: Web UI returns 200 on `/`
- **Build Time**: First deployment takes ~3-5 minutes due to npm dependency installation and TypeScript compilation

## Dependencies for PilotDeck on Railway

PilotDeck is a self-contained application with no external database dependencies. All state is stored in SQLite within the mounted volume.

### Deployment Dependencies

| Service | Image | Port | Volume |
|---------|-------|------|--------|
| pilotdeck | Built from Dockerfile | 3001 | /root/.pilotdeck |

## 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/pilotdeck