The project employs artificial intelligence not as a postproduction tool but as a catalyst for design discovery
Conceptual Approach
From the earliest stages of material research, custom diffusion models generate high-resolution studies of ceramic glazes -testing how color, porosity, and surface texture respond to Stuttgart’s humidity and shifting light throughout the year. These AI-driven images serve as analytical sketches, expanding the scope of experimentation.
The visuals also structure the design narrative. The most evocative frames are carefully curated, annotated with environmental data, and assembled into storyboard sheets that trace a resident’s journey from street to sky court. Each storyboard then acts as a directive for the parametric model: balcony depths, façade perforations, and communal terraces are dimensioned directly from the atmospheres depicted in the imagery. This iterative loop -generate, narrate, parametrize- transforms abstract research into spatial sequences that can be costed, constructed, and lived in.
To move from still frames to immersive previews, the workflow integrates a hybrid pipeline: diffusion models for rapid variation and enhance-focused AI for precise upscaling, ensuring clarity at every stage. The outcome is an architecture that is simultaneously data-driven and story-infused, where every ceramic tile and each corner loggia emerge from a dialogue between human intuition and machine imagination.

Prompts & Software
An apartment made of a green ceramics’ façade. Contemporary architectural style.
Created with Freepik, Topaz Labs and Photo AI.






