Introducing Matry
The practice of design, along with many other disciplines, is experiencing turbulence in the current moment. Regardless of how you feel about AI, it’s clearly here to stay. There’s no shortage of predictions about where this all leads, and certainly no shortage of confidence from those offering up their supposed foresight.
I feel less confident. The past few years have humbled me, such that I’m now wary of trusting my own instincts. I have no idea what the future holds for any of us.
Fortunately, this doesn’t have to be a dismal sentiment. I’m actually more excited to be in this field right now than I ever have. The uncertainty leaves the possibility space wide open, which means we have the opportunity to be weird.
It is in this spirit of weirdness that I introduce Matry.
I came up with the name in 2015, and while I won’t bore you with the story, suffice to say that the concept has undergone dozens of iterations. At its core, the idea was always to turn language into a design tool by offering a textual interface for editing GUIs.
Ever since AI began taking over the tech scene, I’ve been trying to figure out where Matry fits in. LLMs are a pristine expression of Matry’s core idea, so what value could we provide that wasn’t already satisfied by ChatGPT, Claude, and others?
notes:
what do I want to say here?
- designers should be designing in the target medium, which for web means the browser
- a web page in a browser is too constrained an environment
- plugins are finicky and limited
- design tools should assume AI
- you could use CC, but should the future of design work really be in the terminal?
- you could use Cursor, but should design tools be embedded inside a code editor?
- the vision is an AI-native design tool that has unprecedented access to the target environment: the browser