Dataharvest 2025 - the European Investigative Journalism Conference

Investigative journalism often involves analyzing vast amounts of documents. In this session, we will demonstrate how Retrieval-Augmented Generation (RAG) systems can help you getting information out of document collections.

We’ll showcase both open-source solutions and a custom-built tools, using real-world examples from recent collaborative investigations. Whether you're dealing with leaked documents, court records, or government reports, this session will equip you with techniques to get insights from your document faster than using an intern.

After attending this session, you will be able to integrate these tools into your investigative workflow, and apply them to large document collections.

No prior knowledge of AI or programming is required

To follow along, bring your laptop!