What We Developed & Project Impact
EGPAF operates in multiple countries, managing a vast amount of critical data on the pediatric HIV/AIDS epidemic. Today, they have an average of 1.7 billion rows in their database. However, accessing and interpreting this information to make fast and strategic decisions remains a challenge.
Given this scenario, our team had a single mission: to accelerate decisions that save lives. In just 20 days, we developed a fully integrated technological solution capable of transforming access to critical information. Inspired by the Retrieval-Augmented Generation (RAG) architecture, the tool allows users to ask questions in natural language and receive precise, contextualized answers extracted directly from the foundation’s databases.
Our approach goes far beyond a simple AI application. The development started with the creation of a robust data warehouse, consolidating and standardizing fragmented data from 16 countries where EGPAF operates. This enabled the structuring of a unified and coherent database, essential for ensuring accurate and agile analysis.
With Microsoft Fabric support, we built a semantic model on this consolidated database, enabling advanced, multidimensional interpretations and the extraction of strategic insights.