How the University of Arizona’s KMap transforms research collaboration with AI-powered discovery on AWS

This article describes how the University of Arizona developed KMap, an AI-powered research discovery platform built on AWS that helps researchers find collaborators and expertise across the institution.

• KMap aggregates data from 7,000+ researchers across 150+ departments and dozens of data sources
• Uses generative AI and retrieval-augmented generation to answer natural language questions about research capabilities
• Interactive visualization maps departments and collaboration networks across campus
• Migrated from mixed on-premises/cloud infrastructure to AWS for scalability and data governance
• Generates 1.2 million annual interactions with 41,000+ users; 76% organic discovery rate
• Enables rapid expert identification; researchers report saving "a week of manual digging" per search
• Built on AWS services: S3, Lambda, OpenSearch, API Gateway, Bedrock, and SageMaker
• Future features include collaboration recommendations and document-based discovery for students

KMap demonstrates how AI-powered institutional knowledge systems can accelerate research collaboration and innovation by connecting researchers across disciplinary silos.