In 2009 the University of Pavia launched PAD – Pavia Archivi Digitali with the idea of creating an archive aimed at collecting digital born memories of writers, journalists and intellectuals. The archive was planned to include files and documents of a different nature, to ensure their long-term preservation and to make them accessible to scholars, in accordance with the author’s provisions. Following the setting up of a first prototype, the project evolved into a sustainable system, capable of handling large amounts of data and keeping them safe. Having established that the products on the market were not able to support the complexity of the project, the decision was made to develop the system locally, ensuring that it would be scalable, flexible, compliant to standards and interoperable with other platforms. Literary digital archives are mainly made of text documents (working papers, drafts at various stages of completeness, unpublished papers, pdf of gone to press volumes), and to a lesser extent of audio and video-recordings, photographs and images. PAD is currently extending the type of files to be treated, described and stored, searching solutions designed to support, host and make available to scholars digital material produced on the web (social networks, websites, blogs) and via mail. So far we have focused on files copied from physical media used by authors (their main computer hard disk, external hard drives, CDs, smartphones and mobiles) as well as from cloud storage systems (Dropbox, Google Drive and so on). Due to the large size of these archives, often including thousands of files, the development team is studying new techniques for data extraction and AI procedures of file recognition. Franco Buffoni's papers are used as a testbed for these developments.