Speaker
Description
How do you modernize a critical legacy system inside one of the world’s most complex scientific institutions — while keeping your users, your stakeholders, and your sanity intact?
At CERN, we’ve been rebuilding our enterprise document management platform — a core system supporting HR, finance, and everyday operations — moving from a large, tightly-coupled Java application to smaller backend services, a modern React web interface, and a workflow engine that coordinates business processes.
But this talk isn’t about chasing shiny tech. It’s about making solid engineering decisions under real constraints: how to split a nearly 30-year-old system into manageable parts that teams can own; how to introduce clearer release and versioning rules so shared components don’t break each other; how to move from “everything talks directly to the database” to safer API-based integration; how to replace an embedded workflow component with a dedicated BPMN platform; and how to redesign a stateful, server-driven UI into a stateless React frontend backed by modular APIs.
This isn’t a silver bullet — it’s a practical story of trade-offs, pitfalls, and patterns that can help you think about modernizing real systems under constraints, even in smaller projects.