In addition, Airbyte has added two new features to PyAirbyte: Airbyte Cloud orchestration and local docker-based orchestration. The former allows users to orchestrate their Cloud-hosted pipelines, while the latter enables the running of connectors that are not Python-native. They have also launched a new Snowflake Cortex destination, which allows users to create their own dedicated vector store within Snowflake. Finally, Airbyte is moving towards releasing Airbyte 1.0, which will include features like resumable full refreshes, automatic detection of dropped records, and no more stuck syncs.
Key takeaways:
- Airbyte is developing a Connector Builder AI that will simplify the process of creating custom connectors by allowing users to provide a link to the API documentation, and the AI will create the connector.
- Improvements to the Connector Builder include support for Record Filters and Schema Normalization, more flexible management of User Inputs and config values, and automatic updates to the CDK version used by the connector.
- PyAirbyte has added two new features: Airbyte Cloud orchestration and local docker-based orchestration. This allows users to orchestrate their Cloud-hosted pipelines as well as locally-running ones, and to run connectors that are not Python-native.
- Airbyte has launched a new Snowflake Cortex destination that allows users to create their own dedicated vector store directly within Snowflake, and is moving towards the release of Airbyte 1.0, which will focus on reliability and data accuracy.