Keboola
End-to-end data operations platform for pipelines, transformations and orchestration
Keboola is smoothing the seams — MCP login, direct workspace writes, and BigQuery catching up to Snowflake
◆Recent moves
- 8d ago
Keboola MCP: Log In Once, Work Across Every Project
Connecting an assistant to the MCP server now takes a browser login rather than a pasted Storage API token, and one session reaches every project on the stack. Removes the main friction in an MCP server that already existed.
View source ↗ - 13d ago
Python 3.10 Deprecation for Streamlit Data Apps
Streamlit data apps on Python 3.10 must move to a newer backend before the 24 September default switch, after which the old image stops getting fixes. A dated migration users have to act on.
View source ↗ - 23d ago
Flows: New Condition Operators and Run Selected Tasks
Flows gain continue-on-failure, conditions applied across all phases at once, and the ability to run a subset of tasks from the builder. Practical control for people debugging long flows.
View source ↗ - 26d ago
Branched Storage on BigQuery
Branched storage comes to all BigQuery projects, matching what Snowflake projects already had, and is explicitly groundwork for Branches 2.0. Parity work rather than new capability.
View source ↗ - 27d ago
Workspaces Now Write Directly to Storage Tables
Snowflake and BigQuery workspaces can write rows directly into Storage with no output-mapping job or staging step. Cuts a required intermediate out of the common path.
View source ↗ - 28d ago
New Python Versions Available for Transformations and Workspaces
Python 3.11 through 3.14 become available for transformations and workspaces, with existing transformations left on their current version and no default changed.
View source ↗