A locally hosted, open-weight GenAI system built on standard hardware — demonstrating that domain-specific educational AI is feasible for resource-constrained institutions.
Total hardware investment of approximately €5,000 — ensuring data sovereignty, localised control, and independence from commercial cloud services.
The Retrieval-Augmented Generation pipeline actively queries a curated document repository at inference time, constraining outputs within expert-verified knowledge boundaries.
The choice of a locally hosted, open-weight model ensures data sovereignty, provides explicit control over the knowledge boundary, enables transparency of the retrieval process for expert auditing, and eliminates dependency on commercial cloud services — consistent with EU Ethics Guidelines for Trustworthy AI.