Enterprise lab-in-the-loop on AWS: How Sanofi is compressing drug discovery from years to weeks

This article describes how Sanofi built SWEL (Scientific Workflow Experience Labs) on AWS to compress drug discovery timelines by providing AI systems with shared context and continuity across the discovery lifecycle.

• SWEL integrates data, workflows, models, and AI agents into a closed-loop design-make-test-analyze (DMTA) process
• Built on SIPS data layer (20+ petabytes) using Amazon S3, with orchestration via AWS Batch, Amazon EKS, and Amazon MWAA
• SWEL Copilot provides scientists natural language interface powered by Amazon Bedrock AgentCore with governance via IAM and Bedrock Guardrails
• Supports small molecules, large molecules, mRNA, vaccines, and CMC workflows on unified execution plane
• Early projects show improved hit rates; platform deployed in 2.5 months and now supports 50+ scientific workflows
• Context propagates through design, experimentation, orchestration, and reasoning layers to enable autonomous discovery

SWEL demonstrates how context-aware AI infrastructure can transform pharmaceutical R&D by enabling iterative, data-driven discovery at enterprise scale.