The process of building a personal application with AI tools fosters a sense of "permissionless building." This newfound confidence directly impacts professional work, encouraging designers to take more risks, proactively fill gaps, and build solutions without waiting for a formal brief.
As AI makes high-quality execution accessible to everyone, 'craft' and 'quality' will cease to be primary differentiators. The future of design will be defined by 'soul'鈥攖he unique, personal, and human elements infused into the work, moving away from generic templates and trends.
To implement advanced features beyond your expertise, like metal shaders, create a custom AI "skill." You can train an AI agent by feeding it information from various sources, such as YouTube tutorial transcripts and technical articles, compiling a specialized knowledge base for it to draw upon.
Instead of viewing 'I don't know' as a roadblock, use it as a prompt. By admitting your knowledge gaps to an AI like Claude, you invite it to become a co-conspirator, guiding you through complex processes like API integration and fueling creative momentum.
While generating products with AI is popular, a massive unlock lies in applying it to unseen internal processes. AI can optimize workflows, improve content design, and perform analysis. These non-product applications can create significant leverage for design teams within larger organizations.
While AI can run tasks autonomously, creatives must stay "in the loop." Avoid simply accepting AI output; instead, provide constant feedback to shape the result until it feels authentically yours. This prevents generic, soulless work and ensures you remain proud of the final product.
The goal of "permissionless building" isn't always a polished product. AI allows you to create highly specific, "janky" apps in hours to solve unique personal problems, like syncing health data across devices. The value is in the immediate utility, not the public-facing design.
If you're unsure where to start with AI, begin with self-diagnosis. Tell the AI your role, describe your daily calendar and tasks, and ask it to identify where it can help. LLMs excel at pattern matching and can reflect back opportunities for automation you might have missed.
If you find yourself using the same complex prompt repeatedly, codify it into a "skill." A skill is a simple markdown file with instructions that the AI can invoke on command. You can even ask the AI to help you build the skill itself, raising the ceiling of its output and making your workflow more efficient.
Use AI to manage its own development tasks. After a brain dump of project goals, have the AI create tickets in a tool like Linear. Then, let the AI work through the tickets and update its own statuses, significantly reducing your mental load and freeing you up for higher-level review.