Deploy and Host Sim AI on Railway

Sim AI is an open-source platform for building and deploying AI agent workflows. It provides a visual, low-code interface powered by Next.js and Bun, enabling users to create complex LLM-based automations with drag-and-drop nodes for agents, functions, knowledge bases, and integrations like Slack, Gmail, and Pinecone. Self-hostable via Docker or NPM, it supports PostgreSQL with pgvector for embeddings and tools like Socket.io for realtime features. (48 words)

About Hosting Sim AI

Hosting Sim AI involves deploying its full-stack application, which includes a Next.js frontend, Bun runtime backend, and PostgreSQL database with pgvector extension for AI embeddings. Start by cloning the GitHub repository (https://github.com/simstudioai/sim) and setting up environment variables for database connections, API keys (e.g., OpenAI, Copilot), and ports. Use Docker Compose for quick setup, pulling necessary images and running migrations via Drizzle ORM. Configure authentication with Better Auth and enable realtime features with Socket.io. On Railway, link your repo, provision a PostgreSQL database, and deploy—Railway handles scaling, builds with Nixpacks or Dockerfiles, and manages environment vars automatically. Monitor via Railway's dashboard for logs and metrics. This self-hosted setup allows customization for production use cases like agentic workflows, ensuring data privacy and integration with external LLMs. (128 words)

Common Use Cases

• Automating customer support workflows by integrating AI agents with Discord, Slack, or email tools to handle queries and triage issues.
• Building RAG-based knowledge retrieval systems using PostgreSQL pgvector for vector search, combined with tools like Pinecone or Notion for document ingestion.
• Creating scheduled AI pipelines, such as daily data analysis with Google Sheets integration or LLM-powered content generation using OpenAI and YouTube transcription.

Dependencies for Sim AI Hosting

• Bun runtime (for fast JavaScript/TypeScript execution)
• PostgreSQL 12+ with pgvector extension (for AI embeddings and vector database)

Deployment Dependencies

• GitHub Repository – Source code and setup instructions
• Railway PostgreSQL – Managed database service
• Docker Compose – For local testing and container orchestration
• Bun Installation – Runtime environment
• pgvector Guide – Extension for PostgreSQL vector support

Implementation Details

To deploy on Railway, ensure your railway.json or Dockerfile specifies Bun as the runtime. Set environment variables like DATABASE_URL (from Railway's PostgreSQL service), OPENAI_API_KEY, and COPILOT_API_KEY. Run migrations post-deploy with bun run db:migrate. For realtime features, expose port 3000 and configure Socket.io. Example Railway CLI deploy command:

railway login
railway link 
railway up

Link the database service via Railway's dashboard for automatic connection.

Why Deploy Sim AI 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 Sim AI 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.