From Representation to Reasoning: Architectural Intelligence for Adaptability and Circularity

You are cordially invited to attend the lecture by Mikhael Johanes:

Date: 30 Oct 2026
Time: 10:00 AM – 11:00 AM
Venue: SDE3 Level 4, LT427

From Representation to Reasoning: Architectural Intelligence for Adaptability and Circularity

What happens when computational systems shift from representing architecture to reasoning about it? Architectural representation has never been neutral; the way spatial, material, and contextual knowledge is structured shapes what can be understood, compared, and transformed. Recent advances in AI and machine learning make this question increasingly consequential. Tracing a trajectory from machine representations of architectural space to computational systems that combine learned reasoning with algorithmic processes, Mikhael explores how such systems can engage with spatial configuration, programme, and materials in constant flux, thereby foregrounding a shift from conceiving architecture as a singular end state to understanding it as a system capable of being adapted, reused, and transformed across multiple lifecycles.

Mikhael Johanes is a Design++ Postdoctoral Fellow at ETH Zurich, jointly based at the Chair of Circular Engineering for Architecture (CEA) and the Computer Vision and Learning Group (VLG). His research examines how artificial intelligence and computational design can support more circular and adaptable architecture. He earned his PhD from EPFL’s Media and Design Lab, where he developed machine-learning methods for representing, analysing, and generating architectural space.