Model Card Writer AI Prompt

Write a comprehensive model card for this production ML model.

Model cards are documentation artifacts that describe a model's intended use, performance characteristics, limitations, and ethical considerations.

Model: {{model_name}}
Owner: {{owner_team}}

1. Model overview:
   - Model name and version
   - Model type: {{model_type}} (e.g. gradient boosted classifier)
   - Purpose: what task does this model solve? One paragraph.
   - Intended users: who uses this model and in what context?
   - Out-of-scope uses: what should this model NOT be used for?

2. Training data:
   - Data sources: where did the training data come from?
   - Time range: what period does the training data cover?
   - Dataset size: number of examples and features
   - Known biases or limitations in the training data
   - Data preprocessing and feature engineering summary

3. Performance:
   - Primary metric and its value on the test set
   - All secondary metrics
   - Performance broken down by key subgroups (age, region, device, etc.)
   - Performance comparison to baseline
   - Confidence: how reliable are these estimates?

4. Limitations and risks:
   - Known failure modes: when does this model perform poorly?
   - Distribution shift sensitivity: how sensitive is performance to input changes?
   - Uncertainty: what does the model not know it does not know?
   - Potential for harm: could this model produce unfair or harmful outcomes for any group?

5. Ethical considerations:
   - Fairness assessment: performance disparity across demographic groups
   - Privacy: does the model encode or memorize sensitive information?
   - Explainability: can individual predictions be explained?

6. Operations:
   - Model version and registry location
   - Serving infrastructure
   - Monitoring in place
   - Retraining frequency and trigger conditions
   - Owner and escalation path

Return: complete model card document in Markdown format.