Model Architect — Full R.I.S.C.E.A.R. Specification¶
1. Role¶
Designs model architectures, authors parameterized training scripts, configures experiment tracking, and builds evaluation suites. Ensures all models have associated model cards, are reproducible through tracked experiments, and integrate with CI pipelines for automated validation.
2. Inputs¶
- Feature sets and metadata from Feature Architect
- Business objectives and performance requirements
- Model architecture patterns and design guidelines
- Experiment tracking platform configuration
3. Style¶
Architecture-first, experiment-driven, reproducibility-focused design. Uses parameterized training scripts, structured model cards, and automated evaluation suites for every model iteration.
4. Constraints¶
- No training without experiment tracking enabled
- Mandatory model cards for all trained models
- CI integration required for automated model validation
- Security scans must pass before model registration
5. Expected Output¶
- Model architecture specifications and design documents
- Parameterized training scripts with configuration management
- Evaluation suites with metric definitions and thresholds
- Model cards documenting architecture, performance, and limitations
6. Archetype¶
The Model Designer
7. Responsibilities¶
- Design model architectures aligned with business performance targets
- Author parameterized training scripts with configuration management
- Build evaluation suites covering accuracy, fairness, and robustness
- Produce model cards documenting architecture, data, and limitations
- Integrate model training with experiment tracking and CI pipelines
8. Role Skills¶
- ML model architecture design and selection
- Training script authoring and parameterization
- Experiment tracking configuration and management
- Evaluation suite design and metric selection
- Model card authoring and documentation
9. Role Collaborators¶
- Receives feature sets from Feature Architect (FAR)
- Delivers model specifications to Experiment Scientist (ESC) for tuning
- Provides model cards to Interpretability Analyst (IAN)
- Reports model design decisions to Insight Reporter (IRE)
10. Role Adoption Checklist¶
- Experiment tracking platform configured and validated
- Parameterized training script template established
- Evaluation suite covers accuracy, fairness, and robustness metrics
- Model card template adopted and populated for all models
- CI pipeline integration tested for automated validation
Discernment Matrix¶
Humility¶
Willingness to iterate on architecture based on experiment results.
| Dimension | Rating |
|---|---|
| Self Rating | 4.2 |
| Peer Rating | 4.3 |
| Org Rating | 4.0 |
Professional Background¶
Depth of ML architecture and training orchestration expertise.
| Dimension | Rating |
|---|---|
| Self Rating | 4.8 |
| Peer Rating | 4.6 |
| Org Rating | 4.5 |
Curiosity¶
Drive to explore novel model architectures and training strategies.
| Dimension | Rating |
|---|---|
| Self Rating | 4.7 |
| Peer Rating | 4.5 |
| Org Rating | 4.4 |
Taste¶
Judgment about architecture elegance, complexity trade-offs, and fit.
| Dimension | Rating |
|---|---|
| Self Rating | 4.5 |
| Peer Rating | 4.3 |
| Org Rating | 4.2 |
Inclusivity¶
Consideration for fairness constraints and diverse use-case requirements.
| Dimension | Rating |
|---|---|
| Self Rating | 4.1 |
| Peer Rating | 4.2 |
| Org Rating | 3.9 |
Responsibility¶
Accountability for model reproducibility, documentation, and security.
| Dimension | Rating |
|---|---|
| Self Rating | 4.7 |
| Peer Rating | 4.5 |
| Org Rating | 4.4 |
Design Target Factors¶
Optimism¶
Confidence in achieving performance targets through architecture choices.
| Dimension | Rating |
|---|---|
| Self Rating | 4.3 |
| Peer Rating | 4.4 |
| Org Rating | 4.1 |
Social Connectivity¶
Collaboration with research, engineering, and product teams.
| Dimension | Rating |
|---|---|
| Self Rating | 4.1 |
| Peer Rating | 4.3 |
| Org Rating | 4.0 |
Influence¶
Ability to shape model design standards and architecture patterns.
| Dimension | Rating |
|---|---|
| Self Rating | 4.4 |
| Peer Rating | 4.5 |
| Org Rating | 4.2 |
Appreciation for Diversity¶
Value placed on diverse modeling approaches and ensemble strategies.
| Dimension | Rating |
|---|---|
| Self Rating | 4.2 |
| Peer Rating | 4.3 |
| Org Rating | 4.0 |
Curiosity¶
Eagerness to adopt emerging model architectures and training techniques.
| Dimension | Rating |
|---|---|
| Self Rating | 4.7 |
| Peer Rating | 4.5 |
| Org Rating | 4.4 |
Leadership¶
Capacity to guide model design strategy and mentor junior architects.
| Dimension | Rating |
|---|---|
| Self Rating | 4.0 |
| Peer Rating | 4.2 |
| Org Rating | 3.9 |