University of Hamburg - to research, to teach, to educate and form, to homepage
Hub of Computing & Data Science
Hub of Computing
& Data Science

Photo: uhh/hcds

Jurisdiction Support for Refugees through Thoughtful Software

The project Jurisdiction Support for Refugees through Thoughtful Software JuST project addresses a highly sensitive and socially relevant challenge: access to legal information and support for refugees in Germany. In light of demographic change, Germany depends on skilled workers, while many migrants arriving in recent years are already qualified or seek to pursue higher education. However, navigating the German legal system remains a major barrier for many.

Legal regulations are complex, advisory services are often limited, costly, and difficult to access due to language barriers. At universities, this gap is frequently bridged by volunteer-based initiatives such as refugee law clinics. While indispensable, these initiatives face structural limitations in capacity and rarely make systematic use of digital tools.

JuST aims to support volunteer legal counselling through digital, privacy-compliant Legal Tech solutions. Importantly, the project does not seek to replace legal expertise or automate legal decision-making. Instead, JuST focuses on legal orientation, information, and structured guidance, ensuring that human advisors remain central to the process while repetitive manual tasks are reduced.

Within the project, a multilingual software prototype is developed that enables both a 24/7 first point of contact without advisors and targeted support for volunteer counsellors. By streamlining workflows, JuST helps ensure that limited – and often voluntary – resources can be used more effectively.

The development takes place in a higher education context and initially focuses on the legal areas of higher education, temporary residence permits, permanent settlement permits, and preventing deportation. The overarching goal is to enable refugees to pursue secure studies and build independent livelihoods. In the long term, the project explores the transferability of JuST to additional legal domains and its potential use in multilingual public-sector advisory services.

The JuST prototype comprises three core functionalities:

  • Legal support through structured operationalization of relevant legal frameworks,
  • Natural language, multilingual interaction via a chatbot-based interface,
  • Automated generation of applications and reports, with final review and control always remaining with the user.


Technically, JuST is based on a Retrieval-Augmented Generation (RAG) architecture enriched with statutory legal texts. It relies on open-source LLaMA language models to ensure transparency, data protection, and controlled deployment. The system is implemented in Python, with a Node.js-based frontend developed for demonstration and evaluation purposes.

The project is being implemented at the University of Hamburg in cooperation with Eva Bittner, Wiebke Judith, Quint Haidar Aly (ACCICE), Martin Semmann & Marten Borchers (Hub of Computing and Data Science), and Benjamin Klinkigt (Auxiliary AI GmbH) and is funded by Hamburg Innovation GmbH as part of the Call for Transfer.