Deploy and Host Ghostlight AI testing on Railway

Ghostlight AI testing is a self-hosted AI companion starter stack for testing personalized companion deployments. It includes the main Ghostlight app, PostgreSQL for structured data, Qdrant for vector memory, and persistent file storage for uploads, media, exports, and generated assets.

About Hosting Ghostlight AI testing

Hosting Ghostlight AI testing on Railway gives you a ready-to-run environment for testing AI companion deployments without manually wiring the full backend stack yourself. The template includes the app service, database, vector memory service, and persistent storage volume. Users can deploy the stack, add their own API keys and Discord settings, then customize companion identity, memory, models, media, and behavior through the app. This setup is intended for testing, development, and preparing customer-ready Ghostlight companion builds.

Common Use Cases

• Testing a personalized AI companion before a customer deployment
• Running a Discord-based companion with memory, media, and profile settings
• Building and validating Ghostlight companion templates with Postgres, Qdrant, and persistent storage

Dependencies for Ghostlight AI testing Hosting

• PostgreSQL database for companion profiles, settings, logs, memories, and structured app data
• Qdrant vector database for semantic memory search and long-term memory retrieval
• Persistent storage volume mounted at /data for uploads, generated media, exports, backups, images, and voice files
• Discord bot token and Discord application credentials
• AI provider API key, such as OpenRouter
• Optional media provider keys for image generation, voice, YouTube, or Second Life features

Deployment Dependencies

• Railway account
• GitHub repository for the Ghostlight app
• Discord Developer Portal for creating a bot token and application credentials
• OpenRouter or another supported LLM provider for companion responses
• Optional Qdrant API key for protected vector database access
• Optional media API providers depending on enabled features

Implementation Details

This template includes the core infrastructure expected by Ghostlight AI testing:

Ghostlight App
Postgres
Qdrant
Persistent app storage at /data
Qdrant storage at /qdrant/storage

Recommended app environment variables:

DATABASE_URL=${{Postgres.DATABASE_URL}}

QDRANT_URL=http://${{Qdrant.RAILWAY_PRIVATE_DOMAIN}}:6333
QDRANT_API_KEY=${{Qdrant.QDRANT__SERVICE__API_KEY}}
QDRANT_COLLECTION_PREFIX=ghostlight

STORAGE_DIR=/data
UPLOAD_DIR=/data/uploads
MEDIA_DIR=/data/media
EXPORT_DIR=/data/exports
BACKUP_DIR=/data/backups
VOICE_DIR=/data/voice
IMAGE_DIR=/data/images

Recommended Qdrant environment variables:

PORT=6333
QDRANT__SERVICE__API_KEY=replace_with_private_key

Why Deploy Ghostlight AI testing 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 Ghostlight AI testing 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.