Building the World's Largest RAG for Knowledge Management @ CVS Health
OVERVIEW
EXPERIENCE | In Person |
---|---|
TYPE | Lightning Talk |
TRACK | Generative AI |
INDUSTRY | Health and Life Sciences, Retail and CPG - Food |
TECHNOLOGIES | GenAI/LLMs |
SKILL LEVEL | Intermediate |
DURATION | 20 min |
DOWNLOAD SESSION SLIDES |
CVS Health has 300,000+ employees and a diverse portfolio of businesses. Managing knowledge across a complex organization is challenging when knowledge is spread across various sources used by thousands of teams. Finding info you need to perform work can be a challenge in a big company. How do we ensure our employees can access the most relevant/reliable information at the right time/place? We set out to solve the knowledge management problem and improve the experience at CVS Health. We'll highlight our approach to building the world's largest RAG system for knowledge management. Many RAG systems have been POCs that fail to scale. We will describe how we use RAG to create a unified and scalable knowledge platform that serves many use cases and BUs, sharing tech and org challenges we faced, how we overcame them, and strategies to tackle this massive problem. This talk aims to inspire and inform other practitioners and researchers interested in applying RAG at scale to their domains and problems.
SESSION SPEAKERS
Erik Widman, Ph.D.
/Lead Director of Machine Learning
CVS Health