A small, reusable eval framework for scoring AI classifiers: blind classification via Claude Code, confusion matrices, ordinal miss-direction, confusable-pattern trap audits.
-
Updated
Jul 25, 2026 - Python
A small, reusable eval framework for scoring AI classifiers: blind classification via Claude Code, confusion matrices, ordinal miss-direction, confusable-pattern trap audits.
AI ticket triage classifier with an eval harness — severity/routing taxonomy for multifamily IoT support, scored against a labeled dataset with a confusion matrix and confusable-pattern audit.
Backend service for automated support ticket triage, confidence-based routing, and audit logging
Your AI vendor reports 85% deflection. This audits what that number survives.
Gmail issue tracker that fetches unread threads, classifies them into categories, summarizes conversations, and stores results in Supabase and Excel for easy tracking.
Read-only Salesforce case ops as MCP tools — auth delegated to the sf CLI, agent-composable support queries
To associate your repository with the support-ops topic, visit your repo's landing page and select "manage topics."