one schema per project · no shared tables
Real PostgreSQL, designed for you
Describe the data model. Backenly creates the tables, relations, indexes and row-level security, in a schema that belongs to your project alone.
In practice
The test of a schema tool is not creating tables, it is changing them once they hold live data. Ask Backenly to "add a comments table, each task can have multiple comments" and it plans the change against the live backend: the new table, the foreign keys, the APIs, and policies consistent with your existing rules, applied as governed steps and verified afterwards. Ask it to "drop the projects table" and it will not run silently. An approval card shows exactly how many live rows are affected and whether the data is recoverable, and nothing happens until a person confirms. Every change, additive or destructive, gets a restore point. Each project's schema is also isolated in its own PostgreSQL namespace, so no query can cross between projects.
What it is
Backenly designs and applies a complete PostgreSQL schema from a description of your data: the tables, columns and types, the foreign keys between them, the indexes and the row-level security policies. No schema design sessions and no hand-written migration files. What you get is ordinary PostgreSQL. Query it with standard SQL, or through the same PostgREST grammar Supabase serves, with embedded resources and the full filter vocabulary. The generation replaces the design work, not the API you already know.
How it works
Backenly identifies the entities in your description and the relationships between them. It selects a PostgreSQL type for each field, creates foreign key constraints to keep references intact, adds indexes for the queries the model implies, and applies row-level security so the data is protected in the database itself rather than in application code.
Why it matters
Getting a schema right takes experience: normalization, the right types, indexes for the real query patterns, and security that holds on every path. Getting a change to a live schema right takes more. Backenly does both as governed work, so a production-grade schema does not depend on having a database specialist on the team.
What you get
Normalized by default
Tables without duplicated data, with referential integrity held by foreign keys and join tables rather than by convention.
The right types
UUIDs for identifiers, timestamps with time zone for dates, JSONB where the data is genuinely flexible, and text where it is not.
Row-level security
Policies in PostgreSQL decide what each user can read and write, so access control does not have to be remembered on every API path.
Safe schema evolution
Describe a change and Backenly applies it behind a restore point, records it, and routes anything destructive to a person first.
Common questions
Yes. Every project is standard PostgreSQL, in its own isolated schema, with pgvector available for embeddings.
Yes. Provision a read-only connection string on demand, and arm a read-write one explicitly from the dashboard, both scoped to your project's schema. Full pg_dump exports are available at any time, and read-only SQL (joins, aggregates, window functions, CTEs, EXPLAIN) runs from the CLI and over MCP. What Backenly governs is structural change: schema mutations go through planned, verified, reversible actions rather than ad-hoc DDL. Your data stays portable and your schema stays accountable.
Your data is preserved across plan changes. Upgrading adds capacity; downgrading keeps your data within the limits of the new plan.
Related features
- Agent-built backendsTables, relations, APIs, auth and access rules, planned as typed actions, applied as one governed change and verified against the running backend.
- REST APIEvery table is served by PostgREST, straight from the PostgreSQL catalog. Nothing to regenerate: a table created a second ago is queryable now.
- AuthenticationSign-up, sign-in, sessions and access control on every plan, signed with a secret that belongs to your project alone.
- Operations and autonomyNo Docker, no Kubernetes, no cloud console. Backenly runs the backend, snapshots before every deploy, and watches it every minute once it is live.
Try it on a live project
One free project, no card. Connect your agent over MCP and read the verification evidence yourself.