The new Airbnb AI will make it easier for the company’s interface designers to create product source code.
Airbnb, the popular online marketplace and hospitality service company that allows travelers to find suitable places to stay-in during short or long term vacations, has stepped up its game. The company just introduced the Airbnb AI which, according to them, will generate the necessary source code using even basic design sketches.
The company is investing in a machine learning platform that will recognize sketches or drawings and turn them into actionable code readily. Just think of all of the time and effort that will be saved using such a technology.
The Airbnb AI is still under development, but a demonstration from the team who’s working on it already showed the platform’s potential. This latest move from the hospitality service company is a clear indication that more and more companies are starting to bet on artificial intelligence to automate their systems.#Airbnb #AI could be the next big thing in product development and design!Click To Tweet
The Airbnb AI: A way to Streamline the Design Process
In a blog published on the Airbnb Design website, the company’s design site managed by a team of designers, researchers, writers, and others who are working together ‘behind the scenes’ of Airbnb, they explained the cons of the design process and how AI could potentially improve it.
“As it stands now, every step in the design process and every artifact produced is a dead end,” wrote Benjamin Wilkins, Design Technology Lead at Airbnb.
Wilkins went on to explain that work stops whenever one discipline finishes a portion of the project and the responsibility should now be passed to another discipline.
“Projects progress from stakeholder meetings to design to engineering; requirements become explorations, explorations become mockups and prototypes, and these are handed off to developers to become final products,” he said.
However, Wilkins also emphasized that at the core of these cumbersome steps lies “a translation of shared meaning to a different medium in progression toward a common goal, with skilled experts in each domain acting as translators.”
Still, they wanted the Airbnb AI to simplify the “cumbersome steps” of the overall product development routine. Apparently, the best natural point, to begin with, is the design sketching stage.
The Airbnb AI: Turning Drawing Board Ideas to Product Codes
According to Wilkins, they started with sketching to test the possibility of translating their drawings to finished products immediately–a move that could bypass several stages in the development cycle.
“Sketching seemed like the natural place to start. As interface designers, sketching is an intuitive method of expressing a concept. We wanted to see how it might look to skip a few steps in the product development lifecycle and instantly translate our sketches into a finished product,” Wilkins said.
The team began with a working theory that if machine learning algorithms could classify a complex set of thousands of handwritten symbols accurately, it would enable them to classify the 150 components within their system and teach the machine to recognize them.
The Airbnb interface design team then proceeded with building their initial prototype using a dozen hand-drawn components as training data, open source machine learning algorithms, and a small amount of intermediary code to shown some components from the design system into the browser.
During the demonstration, the system already showed massive potential. The team tried using the same AI technology to do the following:
- to live-code prototypes from whiteboard drawings
- to translate high-fidelity mocks into component specifications for the company engineers
- to translate production code into design files for iteration by their designers
According to Wilkins, they believe that artificial intelligence-assisted design and development would be included in the next generation of toolsets.