Surfaice reposted this
AI for repeatable construction. Repeat YC founder (Mighty Buildings, Socap.ai). Former VC with 2 unicorns from seed portfolio. Raised over $180M 💥
When I started Mighty Buildings 7 years ago, we adopted many principles from the automotive industry. But were we right in this? We aimed to approach home construction as a standardized design-to-production process, automating all repeatable features of homes. However, we discovered that developers demand significant customization, making it challenging to achieve standardization at scale without compromising their requirements. We ended up doing over 300 test-fits in 2 years for “standard“ floor plans. Now, when I work on Surfaice, I deal a lot with retail and QSR (Quick Service Restaurant) brands, and I really their approach to store design as a standardized program from Day One. In the industry, we even refer to these standard designs as "prototypes." Since this sector seems to be much closer on standardization to the automotive industry, can we adopt some of those concepts to build AI automation stack to launch new stores quicker? Can you share your comment please? I would love to hear from my network 👇 #AIinconstruction #Futureofretail #AgenticAI #AIworkflows
Slava Solonitsyn Been there, done that with Iqos. For global retail, the building's envelope is one third of the solution. Another third is solvable with a catalog of furniture in a kit-of-part manner, to enable refits in existing real estate. The third third is even further company-specific : it's the required allocated space for marcom collateral.
If we look back at the history of construction in the second half of the 20th century, another compelling example of standardization and unification can be found in the multifamily residential buildings of the Eastern Bloc and Western Europe. These buildings were constructed with a limited set of prefabricated concrete components, which, although sometimes compromising aesthetics and usability, enabled the mass production of affordable housing after WWII.
Cars are standard products made from custom parts, buildings are mostly custom products made from standard parts (4"x2", 8'x4' sheets). Change the parts catalogue and standarize the product is the way to go. #lookatchina
Love this! Having spent years in store development, designing prototypes and executing layouts, I’ve seen how powerful standardization can be in retail and QSR. It allows for both speed and consistency, but with room to adapt for unique brand needs. Bringing this approach to construction through AI could really transform the industry’s ability to scale efficiently.
I hope not. We certainly need less of this form of prototyping.
Slava Solonitsyn, interesting topic! I saw a similar push from the prefab/modular space at KATERRA, where we also tried to blend standardization with the flexibility developers wanted—retail might indeed be closer to achieving that balance.
Leading Provider of Sage Intacct Construction
4wI would think there should be a lot of relevance in using AI to process the data QSR collects from day-to-day in-store operations to then inform new prototypes. Just the other day I was at a QSR and noticed their bins were automated and I thought to myself "I wonder what data that's collecting". The smart brands who want to stay ahead and be in-sync with consumer demands are going to be utilizing this to inform their spaces. When their physical space is competing with the supercomputer in our pockets, they have to be, right?