dbt Labs has announced Fusion, a complete rewrite of their data transformation engine in Rust that maintains the familiar dbt framework while delivering significant performance improvements. Fusion features native SQL understanding across multiple database dialects and promises faster execution than the Python-based dbt Core. The project uses a hybrid licensing model with open source, source-available, and proprietary components - core code is published on GitHub while some advanced features remain exclusive to paying cloud customers. This represents the latest example of data infrastructure companies leveraging Rust for performance-critical workloads while preserving developer-friendly interfaces.
