Instead of getting bogged down in production constraints like failing tests, designers are encouraged to use code to render the most desirable version of an idea. The prototype's value is in communicating the full vision to engineering, not in being a mergeable pull request.
With AI tools that allow natural language querying of business data, designers no longer need SQL to understand user behavior. This democratized access empowers them to contribute to strategy and become holistic product thinkers, not just visual executors.
Designers use AI tools like Claude Code to connect directly to production data sets. This allows them to build realistic, interactive prototypes that challenge preconceived technical limitations and demonstrate the viability of new product directions without deep engineering support.
As designers embraced AI-assisted coding, they found it easier and more powerful to prototype directly in native languages like Swift. This grassroots movement is forcing a strategic conversation at the CTO level to reconsider the company's commitment to React Native.
As designers increasingly use AI to generate and refine work, their ability to provide precise, articulate verbal feedback becomes paramount. The language used in critique is now the direct input for the agent, making a strong design vocabulary more critical than ever.
About once a year, the design team takes 2-4 weeks for "shoplifting," a focused effort to horizontally improve every product surface. This process, detached from the regular roadmap, generates signature, delightful features that might not survive a typical prioritization process.
The idea of setting a yearly vision is outdated when new, compelling prototypes can be generated weekly. At Shopify, strategy now emerges organically as a powerful prototype gets shared, generates excitement, and a team forms around it, shifting priorities in near real-time.
Initial prototypes for the Shop app's less-dense, "window shopping" feed made PMs and data scientists nervous as it thwarted established best practices. Leadership's willingness to build and test the vision-led bet, despite internal friction, was key to its success.
When faced with a blocker from another function, like an engineering constraint or a past data finding, designers shouldn't accept it at face value. With new AI tools, they can independently query data or prototype solutions to challenge these assumptions and form their own perspective.