Provision disposable SQL databases, instantly
Instantly access fresh, isolated databases for CI pipelines, testing and AI workflows.
No setup, or cleanup required.
From the creators of DB Fiddle
We're opening up the battle-tested infrastructure behind DB Fiddle for teams that need ephemeral database environments as a core building block.
Platform capabilities
Built for modern development and AI workflows
Get a clean, untainted database for every workflow instantly, with none of the overhead of managing persistent environments.
- Instantly available
Access isolated, ephemeral databases instantly so every workflow can start from a clean state.
- Automatic lifecycle management
Databases are automatically cleaned up after a configurable duration, or optionally after each request.
- Multi-engine support
Run workloads against a variety of SQL engines including PostgreSQL, MySQL, MariaDB and SQLite.
- Stateless by design
No long-lived infrastructure to babysit. Databases expire automatically after each task or after the configured duration.
- Built for automation
Integrate with CI pipelines, test runners, sandboxes, AI workflows, and internal developer tools.
- Command line or API
Create and manage databases using our CLI or API, whichever is best suited for your workflow.
Example use cases
Designed for deterministic workflows where every run should start with a clean, isolated database.
- CI pipelines.
- A fresh database for every pipeline run - clean, isolated, and instantly available.
- Test runners.
- Spin up real databases for integration tests without setup or shared state.
- Sandboxes.
- Create disposable databases for safe experimentation and query testing.
- AI workflows.
- Give each agent or job its own temporary database for state and processing.
- Internal developer tools.
- Provision on-demand databases via API for scripts, tools, and automation.
- Migration testing.
- Validate schema changes in a clean database before every deploy.
- Data processing jobs.
- Run transformations or imports in isolated databases, then discard them.
- Ephemeral backends.
- Spin up short-lived databases to support temporary services, scripts, or workflows.
- And more...
- The possibilities are endless. Get in touch to discuss your specific use cases.
Sign up now for early access
Join the waitlist to sign up for early access, or reach out to learn more.
Frequently asked questions
Quick answers about the platform and whether it fits your workload. If you have any other questions, please get in touch.
What is a disposable database?
A short-lived, isolated database instance that is created on demand for a task and automatically discarded when finished.
How fast can a database be provisioned?
Provisioning is designed to be instant, with databases typically ready to use in less than a second.
Which use cases are a good fit?
There are many great use cases for disposable databases, including CI pipelines, test runners, sandboxes, AI workflows, and internal developer tools.
How is this related to DB Fiddle?
It uses the same proprietary database infrastructure that powers DB Fiddle, packaged as a developer-facing service.
Is this managed or self-hosted?
The initial offering is fully managed so teams can adopt disposable databases without operating database fleets.
Which database engines are supported?
We currently support a variety of SQL engines, including a number of versions of PostgreSQL, MySQL, MariaDB and SQLite.
Do you plan to support NoSQL engines (Redis, MongoDB etc.)?
We are currently focused on SQL databases, but are open to supporting NoSQL databases in the future.
Interested? Let us knowHow do you handle isolation and cleanup?
Each database is isolated and ephemeral, with automatic lifecycle cleanup to avoid management overhead and stale state.
When will early access open?
We are onboarding in waves based on fit and infrastructure demand. Waitlist members get first updates as slots open.