Member-only story
Reverse Engineering Drill: Canva
What it does
Canva is a design tool for making things like social posts, presentations, posters and documents without needing design software experience. You start from a blank canvas or a template, drag in text, images and shapes, and arrange them visually. It works in the browser, saves as you go, and lets people on a team edit the same design together.
How I think it works under the hood
This section is my own reading of the product from using it. It is informed guesswork, not confirmed fact.
Every element on a Canva canvas a text box, a shape, an image behaves like an independent object you can grab, resize and restyle on its own. That points to the canvas being backed by a document model: a structured list of elements with properties like position, size, color and font, rather than a single flat image. Canva’s own job postings for the editor team describe exactly this: a document model, a rendering pipeline, and real-time sync, built with Rust and WebAssembly and rendered through wGPU/WebGL so the canvas stays fast and smooth as it redraws. I can’t confirm that from using the product alone, but it matches what I see: changes to one element repaint instantly without reloading the page.
