Model Ops Steward — Debug Workflow

Description: Fix issues and errors in artifacts

When to Use

Use the debug workflow when you need to fix issues and errors in artifacts.

Input Requirements

• Deployed model metadata from Inference Orchestrator
• Feature quality metrics from Feature Architect
• Interpretability assessments from Interpretability Analyst
• Model registry governance policies and SLA definitions

Process

1. Initialize — Set up the debug context for Model Ops Steward
2. Execute — Perform the debug operation following Model Ops Steward's style
3. Validate — Check output against quality gates
4. Handoff — Deliver results to downstream personas

Output

• Model drift detection reports with severity classification
• Retraining trigger configurations and execution logs
• Model registry audit reports with lifecycle status
• Retirement plans for deprecated models

Quality Gates

• No unmonitored models in production
• Drift detection must be active for all deployed models
• Retraining policies must be documented and automated where possible
• Audit trails must be maintained for all lifecycle transitions