Rootcode OÜ – Urbanora

Challenge: Prague: How can the feedback gap between citizens and authorities be narrowed?

Target Group: General Population

This pilot, hosted by the Municipality of Prague, focused on developing an AI-powered urban issue reporting system designed to improve how urban issues are reported and handled. The core objective was to create a backend system that automatically classifies citizen-reported issues—submitted with images and descriptions—and routes them to the appropriate municipal departments. The system is intended for integration with existing platforms.

Co-creation was a central part of the process. Two sessions were conducted: one online with the product manager of an existing platform and OICT representatives using a semi-structured interview format, and a second in-person brainstorming session in Prague with the OICT team. These sessions helped validate the technical approach, shape system features, and align the solution with real-world workflows.

As a result, an open-source issue routing and classification system was developed, featuring automated classification using LLMs and image/text anonymization to meet user privacy standards. The system has been tested in a relevant environment and is ready for further refinement and integration.