Case Study: a global technology company builds richer knowledge graphs with Persistent Systems and Google GenAI

Enabling Richer Data Experiences with Google and Generative AI

Persistent Systems helped a Global Technology Company that manages digital imagery and media enhance its core product by turning entities extracted from unstructured documents like PDFs and video transcripts into a knowledge graph. The customer needed a way to build and visualize relationships between extracted entities so users could get deeper context, analytics, and data-sharing value from their content.

Persistent Systems used Google Gemini 1.0 Pro and Microsoft OpenAI GPT-4, along with its NOVA accelerator, to automate document ingestion into Neo4j, generate entity extraction and embeddings, and support semantic search and complex analysis. The project moved from workshops to ROI analysis in eight weeks, and the production solution was delivered in six months, resulting in about a 70% reduction in manual effort, a 50% increase in data volume, and roughly 40% cost savings.