Stiftung Zentrale Stelle Verpackungsregister

For the Central Packaging Register Foundation, VAGO developed an AI-powered ticket pipeline that automatically classifies, processes, and responds to incoming inquiries. The solution, which is operated entirely on-premises, relieves employees of repetitive tasks and integrates seamlessly into the existing Jira infrastructure. By using resource-efficient models on off-the-shelf GPUs, the system achieves a classification accuracy of over 90 percent and frees up capacity for more technically demanding tasks.

SPARK Project As part of the SPARK project, VAGO Solutions, in collaboration with PricewaterhouseCoopers, has developed AI-powered open-source modules designed to assist regulatory agencies in reviewing extensive application documents. The solution automatically analyzes documents, checks them for completeness and plausibility, and prepares structured review recommendations for case workers. Throughout this process, the responsibility for decision-making remains with humans at all times. The agent-based assistance solution is made available as open source via openCode and increases efficiency in complex administrative procedures without compromising transparency or control.

ESA Projekt ac-URBAN

As part of the ESA ac-URBAN project, VAGO Solutions is developing the AI components for an open-source framework that makes Earth observation data accessible via natural language. The system combines large language models with geospatial data processing, enabling municipalities, planning firms, and environmental organizations to analyze complex satellite data without in-depth expertise in geospatial data. User queries are automatically translated into reproducible analysis workflows, and the results are presented as interactive maps, analyses, and reports. The fully open-source architecture enables transparent analyses for urban climate adaptation, infrastructure planning, and environmental monitoring.