Gastroenterologist Dr. Shaun Chuah developed Foundry120, a research operating system built with Django that leverages agentic AI to manage 30,000 biological samples and 10 terabytes of clinical research data.

Key Points

• Foundry120 uses Django for secure, role-based access control and Azure Blob Storage to centralize diverse datasets like genomics and endoscopy videos.
• The platform integrates Helix, an agentic AI assistant that autonomously writes and executes Python code to analyze data within isolated, secure virtual machine sandboxes.
• Dr. Chuah built the system to replace manual, error-prone tracking in Microsoft Excel, enabling researchers to perform complex data joins and analysis in minutes.
• The architecture prioritizes data governance and security by keeping all inference and compute within the University of Glasgow’s compliant Azure cloud environment.
• The platform supports high-performance, parallelized bioinformatics pipelines, utilizing hundreds of Azure CPUs to process large-scale datasets on demand.

Why it Matters

• This project demonstrates how domain experts can build sophisticated, production-grade research infrastructure by leveraging battle-tested frameworks and cloud-native tools. It highlights a shift in scientific workflows where the researcher's role evolves from manual data processing to supervising and verifying AI-driven analysis.