Etienne Juliot

Co-founder  and currently Vice President of Sales at OBEO

From Natural Language to System Architecture: Combining AI and MBSE for Infrastructure Engineering

Bio.
Etienne Juliot : Co-founder of Obeo, he is currently Vice President of Sales.With a Master’s degree in software architecture from the University of Nantes, Étienne Juliot has more than 25 years of experience in software engineering and model-driven engineering (MDE).Very involved in several Open Source communities such as Sirius, Capella and SysON, he lectures all over the world and served on the Eclipse Foundation’s Board of Directors for fifteen years.

Abstract.

Infrastructure systems are complex by nature. They involve many components, interfaces, constraints, stakeholders, and large amounts of information often scattered across documents and expressed in natural language.

Generative AI offers new possibilities for navigating this complexity. But generating engineering content is not enough. To support actual engineering decisions, information needs to be structured, connected, traceable, and captured within a consistent system architecture.

In this talk, we explore how AI and Model-Based Systems Engineering can complement each other to support the modeling of infrastructure architectures. Through a practical use case, we will show how AI can help engineers extract relevant information, identify architectural concepts and relationships, and accelerate the creation and exploration of a system model.

We will also discuss where the model becomes essential: providing an explicit architecture, maintaining consistency between engineering concepts, and giving engineers a reliable foundation on which AI-assisted workflows can operate.

Rather than replacing systems engineering, AI can make MBSE more accessible and help engineers move faster from unstructured information to an explicit, navigable system architecture.