In this talk, I present ZEW’s rollout of our internal AI frontend based on OpenWebUI, which integrates local and cloud LLMs into researchers’ daily workflows. I will give an overview of our technical setup, including the use of MCP-tools to provide models with useful context, code templates, and research utilities. The talk also highlights our first generation of domain-specific agentic tools, ranging from lightweight task-specific agents, such as an automated public tender screening assistant, to custom tools for coding support and semi-automated data preparation. I will conclude with early lessons from deployment and researcher adoption.
Analyses, survey insights and a practical guide mark a key milestone in the Erasmus+ project supporting data-driven vocational training across five European countries.










