Skip to content

knowledge_base: add ml_experiments_and_monitoring example - #187

Open
andresgarciaf wants to merge 3 commits into
databricks:mainfrom
andresgarciaf:trio/kb-8-ml-experiments-monitoring
Open

andresgarciaf wants to merge 3 commits into
databricks:mainfrom
andresgarciaf:trio/kb-8-ml-experiments-monitoring

Conversation

@andresgarciaf

Copy link
Copy Markdown

Summary

  • MLflow experiment and a Unity Catalog registered model with EXECUTE grants for a consumers group.
  • Serverless training job that sets the champion alias and writes an inference table, run once on deploy via job_runs.
  • Lakehouse Monitoring quality monitor with a daily refresh schedule (off in dev) and failure email notifications.

Testing

Command Result
databricks bundle validate OK
databricks bundle deploy Success
databricks bundle run Success
databricks bundle destroy Success

Notes

  • bundle destroy reports the registered model as deleted but keeps it while model versions exist; the README deletes versions first.

andresgarciaf and others added 3 commits October 22, 2026 00:55
Adds an MLflow experiment, a Unity Catalog registered model with
EXECUTE grants for a consumers group, a serverless training job that
sets the "champion" alias and writes an inference table, a deploy-time
job run, and a Lakehouse Monitoring quality monitor with a daily
refresh schedule and failure notifications.

Co-authored-by: Isaac <no-reply@databricks.com>
Co-authored-by: omnigent <noreply@omnigent.ai>
…ceholder

Auth now comes from the CLI profile. README step 1 only sets the
bundle variables.

Co-authored-by: Isaac <no-reply@databricks.com>
Co-authored-by: omnigent <noreply@omnigent.ai>
Move the serverless job environment from version 4 to version 6,
the newest GA version.

Co-authored-by: Isaac <no-reply@databricks.com>
Co-authored-by: omnigent <noreply@omnigent.ai>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant