databricks Databricks Release Notes ·

Databricks MLflow Trace Explorer Now Supports Custom Views

mldatabrickspreviewdata-scientistdatabricks-mlflow
feature

The Databricks MLflow Trace Explorer now enables users to create custom trace views. Users can describe desired layouts in plain language, allowing the 'Genie' feature to generate reusable views that include relevant trace fields, metrics, and feedback controls. This enhancement is designed to streamline review workflows for machine learning development. This capability is currently available in Beta.

Features (1)
  • Build custom trace views in MLflow Trace Explorer

    Users can now define custom trace view layouts using plain language, with the 'Genie' feature generating reusable views. These views surface relevant trace fields, metrics, and feedback controls for streamlined review workflows, enhancing the debugging and analysis process.

Read the original announcement →

https://docs.databricks.com/aws/en/release-notes/product/2026/august#custom-trace-views-in-the-mlflow-trace-explorer-are-in-beta

Related releases