Data Packages
Connect. Select. Query.
Data packages are read replicas of datasets in your data warehouse you can query with ultra low latency. Select tables and start coding against high performance APIs instantly.
Data packages — read replicas with a twist
Publish effortlessly.
Connect a data source and publish APIs with support for low latency analytics, search, and lookups.
Build confidently.
Query interfaces are backed with a smart cache that insulates you from breaking schema changes.
Query instantly.
Import a data package into your project and run queries without writing boilerplate, waiting on credentials, or adding load on your warehouse.
Works with all your data
Caio Pedroso — Data Engineer
How it works
Integrate data warehouse datasets into your product
Publish data as APIs and code packages.
Select tables or dbt models from a source and publish query interfaces instantly. No ETL or API development necessary.
Configurable runtime interfaces like Python, TypeScript, C#, and GraphQL.
Contextual metadata provides descriptions, ownership, and schema definitions.
Query data immediately.
Run OLAP, OLTP, and search queries with a single interface. Data packages are immediately queryable and designed for seamless integration in application code.
Federate into a larger GraphQL API or import as a library into your code.
The generated query interface provides helper functions and autocorrect in your IDE.
Optimize when you’re ready.
By default, queries hit your compute engine. When moving into production, upgrade the package for low latency and high availability across all workloads.
Accelerate packages and guarantee reliability with one click.
Switch a data package’s source or other configurations with GUI or code.
