Parking structures… the things they paved paradise with. These hulking concrete giants dominate suburban landscapes, their cavernous interiors echoing with the lonely hum of idling engines. These usually uninspiring pieces of infrastructure were once critical for feeding society’s insatiable hunger for car space. But times are changing and adaptive reuse architects are now faced with an intriguing challenge: What can we do to repurpose parking structures when they become obsolete?

As cities rediscover public transit and walkable environments, while innovating with ride-sharing and eventually autonomous vehicles that don’t need to spend 90% of their existence sitting in parking spots taking up space, these concrete leviathans will eventually stand largely unused as monuments to a bygone era. Yet, demolition isn’t the only answer. Could these inhuman structures be the seeds for a new human-centric urban environment? Let’s find out by exploring how the design challenges of parking structure adaptive reuse and learning how architects and their AI tools can help us design a parking structure retrofit for the modern era.  

This post will be the first in a series of posts called “Style Bytes”, each exploring a style I have been working with lately, from the familiar, to the novel. Each post will explain the core design concept, list key features, and provide some examples of the style applied to different contexts. 

I predict that exercises like Style Bytes, where architects use AI style references to evolve one idea into an entire family of design concepts, will become a key part architectural practice as AI tools become more widely adopted. 

Furthermore, platforms for sharing, discussing, and remixing these ideas will provide the constructive criticism and consensus-building needed to turn the firehose of random AI generated content into actual focused and feasible design movements as communities coalesce around entirely new design ideas for the 21st century. 

So, let’s take a look at our first Style Byte, Sunrise Prairie.

Alpaca (a.k.a. Chroma by AlpacaML) is a free AI image generator and AI rendering tool currently in open access with some impressive capabilities for architects, especially for rendering and refining your building concepts. At the time of writing, it offers 100 free generations a day, with reasonably priced options for standard and pro plans, but for most architects looking to experiment and learn AI render tools, the free tier is plenty. You can get started experimenting with Alpaca for free at AlpacaML.com.

One of the issues with many competing AI models is a lack of authorship (i.e. how much control you have over the results compared to the algorithm). A platform with poor authorship means the algorithm makes most of the decisions for you and while its results might look good, it doesn’t take direction all that well and you often can’t reliably edit the generations you get.

Alpaca fixes this issue. Its sketch-to-render and editing tools let you turn everything from well-developed Rhino models to even the worst Microsoft Paint doodles into stunning images. It also does this with a fast generation time, giving you constant feedback so you can refine the result by refining your drawing. This gives a tremendous amount of authorship to the user and makes fine tuning images a breeze compared to the uncontrollable slot machine that platforms like Midjourney provide (still love you though Midjourney).