The
AexM
Blog

Introducing the Architecture ex Machina Forum

AexM Team - 18 April 2019

We're excited to introduce the Architecture ex Machina forum! As we seek to engage professionals from across the industry, we wanted to create a place where ideas and conversations could easily be shared. The new forum is powered by Discourse to make it as easy as possible to build a thriving and constructive community.

Differentiable Drawing

Tyler Kvochick - 7 February 2019

Harvard Business Review and Autodesk recently published a white paper on the emerging practice of generative design. In it, they write: 'Manufacturers of all sizes can use advanced generative design tools, which employ algorithms to transform a designer’s requirements into product geometry, optimizing the product design based on the conditions and constraints provided. The tools can provide a plethora of design options in the amount of time it would take engineers to set up a meeting to discuss just one.' While this sounds promising, it is too vague to explain what is happening inside these “generative design tools” that makes them different in a meaningful way. Software companies continue this trend of making vague, non-technical promises that began with the incorporation of software into all industries around 40 years ago. Since then, professionals of all disciplines have gained software tools with the promise of new ways of working, but without an understanding of how the tools work.

A Turing Test for Design

Tyler Kvochick - 17 January 2019

Architecture ex Machina is a group of architecture, engineering, and construction (AEC) professionals dedicated to advancing the state of the art in applications of machine learning to the design of buildings and cities. Although computer aided design (CAD) and building information modeling (BIM) have transformed the design process for buildings, we still have not found a way to capture and interrogate the design process itself. We think that technologies built on machine learning will fundamentally change this process by allowing us to model generative functions and statistical relationships in addition to the drawings and digital models of the built environment.