Strange Loop

Next: September 12-14 2019


Stifel Theatre


St. Louis, MO

Program synthesis for declarative building design

Architects and designers are increasingly embracing scripting paradigms to generate complex and beautiful forms that are too complicated to draw by hand, but these ideas are still fairly new. We are using program synthesis techniques to bridge the gap between programmatic and direct modeling tools, making parametric and constraint-based modeling accessible to a wider audience.

One of the core technologies supporting this effort is Satisfiability Modulo Theories (SMT) solvers. SMT solvers grew out of SAT solvers in the early 2000s and have seen application in software and hardware verification. In this talk, we will walk through our experiences adapting the technology to the radically new domain of architecture and build an example system around an SMT solver.

Drew Wolpert

Drew likes working on systems that help people understand complicated problems and climbing colorful rocks. She got excited about algorithms and theory of computation as an undergrad at MIT, wrote her first production code at Palantir Technologies, and then learned about architecture applications at Flux Factory.

Andrew Zukoski


Andrew Zukoski writes code. Sometimes some of it is okay. He runs, climbs fake rocks and is a total nerd for programming languages and computational geometry. He got started at Bespoke Innovations and 3D Systems creating 3D printed medical devices. He then moved into the architectural software world at the Flux Factory, where among other things he got everyone on his team to implement a Forth. He is now at Join.