From AI-Assisted Coding to Maintainable Enterprise Software
Renè Lamber, Research Industrial Systems Engineering (RISE) Italia
AI coding tools can get software up and running surprisingly fast. But will another developer still understand, operate and extend it three years from now?
In our December session, Renè Lamber from RISE Italia will explore how AI-assisted development can support not just speed, but long-term maintainability. Using the RISE Italia Dashboard as a real-world example, he’ll take us through the architecture of a modular business application built with Java and Quarkus, React and TypeScript, PostgreSQL, Keycloak, Nextcloud, OpenSearch, Tesseract and other open-source components.
We’ll follow a small feature from specification to deployment and see how agent instructions, reusable skills and OpenSpec connect requirements with implementation, while tests and CI help make changes verifiable. The journey continues through containerisation, Kubernetes, GitOps and Argo CD – with examples taken directly from the existing project repository.
The talk will also look beyond working code to an important distinction: building an enterprise application is not the same as proving that it is operationally ready. Security, backups, recovery and clear ownership still matter – no matter how much AI helps along the way.
See how AI can speed up development without turning your codebase into tomorrow’s maintenance problem!