AngryKaren
An automated evaluation platform that tests customer service representatives using realistic AI-powered caller personas. Is your customer service measuring up to your standards?
Project Description
An automated evaluation platform that tests customer service representatives using realistic AI-powered caller personas.
The system generates configurable bot personalities via ElevenLabs voice agents that place actual calls to company CSRs, simulating real customer interactions across various scenarios (complaints, inquiries, technical issues, etc.). The platform records and analyzes conversations to assess CSR performance on key metrics like response time, problem resolution, empathy, policy adherence, and communication quality—providing companies with objective, scalable QA insights without manual mystery shopping.
Key Features:
- Customizable Personas: Configure caller profiles (frustrated customer, confused first-timer, VIP client, etc.)
Natural Voice Interactions: ElevenLabs voice AI enables realistic phone conversations - Automated Scoring: Evaluates CSR performance on multiple dimensions
- Scalable Testing: Run hundreds of evaluations across teams/locations simultaneously
- Actionable Analytics: Dashboard with performance trends, coaching opportunities, and training gaps
Prior Work
None