domo/data-science

Automating end-to-end ML workflows inside Domo involves unifying data access, model development, and deployment using Jupyter workspaces, AutoML pipelines, and scripting tiles to turn data into actionable insights.

Yes, you can use Jupyter workspaces in Domo for exploratory analysis and model training by utilizing persistent notebooks with Python/R kernel options, dataset attachments, and scheduled runs.

Domo supports AutoML and AI service calls to enable no-code, model-assisted training alongside model deployment, scoring, and explanations directly within your data science workflow.

Running inline scripting within Domo ETL pipelines is best handled using scripting tiles, which allow you to execute custom logic and integrate pre-built accelerator patterns for common analytics scenarios.

Maintaining model quality after deploying ML pipelines in Domo requires using built-in monitoring and retraining triggers to track performance and automatically update models as data shifts.

Domo provides pre-built analytics accelerator patterns for common business scenarios including churn prediction, forecasting, and anomaly detection to rapidly deploy data science insights.