How Regeneron is discovering new treatments with AI

The Regeneron Genetics Center built a database pairing sequenced exomes and EHRs for over 400,000 people, but decentralized genomic/clinical data made analyzing their full 10TB dataset (80+ billion data points) difficult, and legacy architecture was costly to scale, with ETL taking days before analytics could begin. Regeneron moved to a Databricks platform on AWS with automated cluster management and Spark-powered pipelines. Query runtime for full-dataset queries improved roughly 600x, from 30 minutes to 3 seconds, ETL pipelines went from 3 weeks to 2 days, and pipelines overall ran 10x faster. This freed bioinformatics and computational biology teams from infrastructure work to focus on research and treatment discovery.

Databricks is a unified data-and-AI lakehouse platform, used across pharma R&D for everything from early prototyping to production model governance via Unity Catalog.

This sits in the Design & Stress-Test stage of the AI use-case lifecycle: Named examples from the tools pharma teams use to scope, prototype and benchmark an AI use case before committing to build it — data-catalog, sandbox and evaluation platforms.