Video

How Villanova democratizes AI with multimodal models and a low-code framework

In this interview, Fabio Catalano, Innovation and Project Orchestration Manager, presents the main objectives, technical results and benefits of the Villanova project. The initiative combines multilingual and multimodal generative AI models with time-series forecasting capabilities and an open-source low-code/no-code framework for building flexible, composable applications. Catalano explains how natural-language interaction can help businesses, public administrations and non-technical users create and maintain customized AI solutions more easily. The interview also introduces Villanova’s key technical outputs, including a European multilingual language model, a federated fine-tuning pipeline and reusable components supporting applications in tourism, agriculture, legal services, public administration and data analysis. By promoting interoperability and reducing dependence on individual technology providers, Villanova aims to make advanced AI more accessible across the European cloud-edge ecosystem.

Watch the video ↓

 

You might also be interested in

How Villanova turns federated generative AI into real-world use cases

In this interview, Luca La Brocca, Business Development Manager, explains how the Villanova project applies federated generative AI to real-world sectors such as healthcare, tourism and precision agriculture. Unlike isolated AI demonstrations, Villanova develops composable and reusable building blocks designed to operate securely across distributed cloud-edge environments. La Brocca presents a practical tourism use case in which specialized language models analyze online reviews and social media content to identify traveler sentiment and emerging trends, providing destination managers with intelligent decision-support tools. The interview also explores Villanova’s contribution to the European research ecosystem, its focus on data sovereignty, privacy and AI Act compliance, and its medium-term objective of transforming research results into scalable, sustainable and market-ready AI solutions.

How high-quality datasets improve LLM training

In this interview, Paolo Budroni, AI Project Manager, explains why high-quality datasets are essential for reliable LLM training. He discusses the importance of collecting diverse and representative data, removing duplicates, spam and toxic content, and applying normalization, anonymization and source balancing. The interview also challenges the assumption that more data is always better, showing how smaller, carefully curated datasets can outperform larger but noisy collections. Finally, Budroni highlights privacy, fairness and regulatory compliance, including GDPR and the EU AI Act, as fundamental elements of ethical dataset curation and user trust.

How Villanova supports Europe’s sovereign cloud-edge ecosystem

In this interview, Alessio Manca, Solutions Partnership Manager, explains Villanova’s role within the IPCEI-CIS programme and the broader 8ra initiative. The project contributes to the development of an open, interoperable and future-proof cloud-edge infrastructure designed to strengthen European digital sovereignty. Manca discusses Villanova’s leadership in service orchestration within Workstream 3 and its contribution to artificial intelligence, software development, governance and open standards. The interview also presents the project’s main technological outputs, including European-trained large language models, a federated fine-tuning pipeline and a low-code/no-code platform for building composable AI-agent applications. Finally, Manca explains how Villanova can support innovation in Italy across tourism, agriculture, legal services and public administration.