Track DAG runs, trends, and performance directly in the Airflow workspace
This is the first major upgrade to the TensorStax verifiers. TensorStax can now simulate DAGs in memory with full operator and database behavior, no execution required.
Airflow now gets its own full workspace with editing, preview, and version control
Each TensorStax instance now runs its own isolated Vault for managing sensitive credentials
Inline editing is now more accurate for complex dbt models and can apply many more diffs at once
Schedule dbt jobs to run on a recurring cadence directly from the workspace
Git integration is now built into the dbt projects page for full version control
The reasoning agent now builds and updates a persistent global context as it explores your project
The reasoning agent now auto-runs deep research to find the most relevant context for each session
Pull in database DDLs as context via the command K menu
Data Viewer for Snowflake, MySQL, and Postgres
Visualize how models connect across your project with the new dbt lineage graph in the dbt projects page
TensorStax now lets you work on dbt models in full-screen mode with an embedded side chat for inline edits and live result previews
Take over and refine dbt models and Airflow DAGs after the agent generates them
TensorStax now streams live previews of dbt models as table outputs in the UI driven by the agent
TensorStax now lets you create and manage pull requests on GitHub and GitLab for your Airflow DAG changes
TensorStax now embeds a reasoning agent that plans between actions for more granular task execution
TensorStax now enforces your organization’s specific Airflow DAG and dbt model formats, including custom macros and data modeling patterns
TensorStax now lets you develop, validate, and lint dbt models alongside Airflow DAGs with instant feedback
.avif)
.avif)
