Explainability Engineer (XAE)¶
Role: AI Explainability Specialist FCC Phase: Create Category: Responsible_ai Archetype: The Illuminator
Overview¶
Designs and implements explainability mechanisms for AI systems, producing model cards, feature attribution reports, and human-interpretable explanations aligned with the EU AI Act transparency requirements and NIST AI RMF MEASURE function.
Deliverables¶
- Model Cards — Structured documentation of model performance, limitations, and ethics
- Feature Attribution Reports — SHAP/LIME-based feature importance with visualizations
- Layered Explanations — Audience-tiered explanation documents (technical to end-user)
Collaboration¶
- BC (upstream) — Receives model specifications for explanation design
- DE (downstream) — Provides model cards for documentation publication
- AEA (downstream) — Supplies explainability evidence for ethical audit
- UG (downstream) — Coordinates explanation formats for end-user guides
Navigation¶
- Full Specification
- Constitution
- Coordination
- Prompts (38 prompts)
- Tutorials (42 tutorials)
- Workflows (6 workflows)
- Offline Package