Introduction
AetherDB is an AI-native database infrastructure platform — like Supabase or Neon, but built from the ground up for AI applications and agents.
What makes AetherDB different
- Natural language queries — ask questions in plain English, AetherDB generates and executes SQL automatically using Groq
- Multi-tenancy built in — every user gets their own isolated Postgres schema, auto-provisioned on signup
- REST + gRPC — both protocols supported out of the box
- MCP server — AI agents like Claude, Cursor, and Copilot can connect directly
- pgvector — native embedding storage and similarity search
- Realtime — database change streams via SSE, powered by Postgres LISTEN/NOTIFY
Architecture
AetherDB runs as a single Go binary that exposes both a REST API and a gRPC server. Each user gets their own isolated Postgres schema. The AI query engine connects to Groq to convert natural language to SQL.
Your App
└── aetherdb-js SDK
└── https://aetherdb.cloud
├── REST API (:8080)
├── gRPC (:50051)
└── PostgreSQL
├── tenant_1 (user A's schema)
├── tenant_2 (user B's schema)
└── tenant_N ...Hosted vs self-hosted
AetherDB is available as a hosted service at aetherdb.cloud. The source code is open and can be self-hosted on any Linux server.
Next steps
- Quickstart — get up and running in 5 minutes
- JavaScript SDK — install and use aetherdb-js
- API reference — full REST API documentation