Add opt-in telemetry to track model performance, errors, and usage patterns across the library.
Currently we have no visibility into:
- Which errors users encounter most frequently
- Model inference times in production
- Which models/features are most used
This data would help prioritize bug fixes, optimize performance, and guide feature development.
Add opt-in telemetry to track model performance, errors, and usage patterns across the library.
Currently we have no visibility into:
This data would help prioritize bug fixes, optimize performance, and guide feature development.