datarobot-model-monitoring

What problem does it solve?

This Skill helps you continuously monitor deployed DataRobot models so you can detect performance degradation, data drift, and prediction anomalies before they impact users.

Core Features & Use Cases

• Performance monitoring: Track prediction volume, latency, and accuracy-related metrics over time, and compare production metrics to training baselines.
• Data drift detection: Identify feature drift and target drift (when actuals are available) and quantify drift severity for investigation.
• Prediction monitoring: Detect unusual prediction patterns and monitor prediction distribution changes and anomaly signals.
• Model health management: Assess health status, generate monitoring insights, and support alerting and retraining trigger workflows.

Quick Start

Use the skill when you want to check the health of a deployment (for example, deployment abc123) and get a report highlighting any significant drift or anomalies.