Medhanie Yemane

Audi receives high volumes of customer feedback daily—from app store reviews to touchpoints across the customer journey—rich with potential to improve software functions and vehicle features. But systematically evaluating this data at scale remains a key challenge.

In this talk, we present a proof-of-concept built with Camunda that automates the ingestion of app store reviews, applies AI-based clustering, and stores the results in a structured data repository. This foundation powers dashboards and an LLM-based chat interface for exploring customer feedback, trends, and pain points—delivering a working MVP in a short timeframe with clear extensibility toward additional channels and agentic follow-on processes.