Solutions · AI apps
A backend for AI apps with vectors and MCP.
Embeddings next to your data, semantic search with RLS, and an MCP server through which agents work safely on your database.
pgvector built in
HNSW indexes, hybrid search and SQL filtering in a single query.
MCP server
Claude, Cursor and other agents run SQL and deploy Edge Functions — with your role's permissions.
Edge Functions for LLMs
Model API keys kept on the server, responses streamed to the client.
Data privacy
Documents and embeddings stay in the EU — important for legal and healthcare customers.
Included
What you get as ai apps
- RAG without a separate vector database
- AI chat function template (Claude)
- Realtime for streaming
- Personal API tokens
Build in a weekend. Scale to millions.
Spin up a free project in 2 minutes. No credit card, no pausing after a week.