When iterating on a Gemini 3.0-generated app, the host uses the annotation feature to draw directly on the preview to request changes. This visual feedback loop allows for more precise and context-specific design adjustments compared to relying solely on ambiguous text descriptions.

Developers can create sophisticated UI elements, like holographic stickers or bouncy page transitions, without writing code. AI assistants like CloudCode are well-trained on animation libraries and can translate descriptive prompts into polished, custom interactions, a capability many developers assume is beyond current AI.

The handoff between AI generation and manual refinement is a major friction point. Tools like Subframe solve this by allowing users to seamlessly switch between an 'Ask AI' mode for generative tasks and a 'Design' mode for manual, Figma-like adjustments on the same canvas.

While chatbots are an effective entry point, they are limiting for complex creative tasks. The next wave of AI products will feature specialized user interfaces that combine fine-grained, gesture-based controls for professionals with hands-off automation for simpler tasks.

Move beyond basic AI prototyping by exporting your design system into a machine-readable format like JSON. By feeding this into an AI agent, you can generate high-fidelity, on-brand components and code that engineers can use directly, dramatically accelerating the path from idea to implementation.

A custom instruction defines your design system's principles (e.g., spacing, color), but it's most effective when paired with a pre-defined component library (e.g., buttons). The instruction tells the AI *how* to arrange things, while the library provides the consistent building blocks, yielding more coherent results.

To rapidly iterate on interactive ideas in code, create your own version of "Command D." Instead of hard-coding values, build a simple control panel with variables for parameters like speed or distance, allowing for easy adjustment and testing of multiple variations.

Open-ended prompts overwhelm new users who don't know what's possible. A better approach is to productize AI into specific features. Use familiar UI like sliders and dropdowns to gather user intent, which then constructs a complex prompt behind the scenes, making powerful AI accessible without requiring prompt engineering skills.

Instead of building UI elements from scratch, adopt modern libraries like Tailwind's Catalyst or Shad CN. They provide pre-built, accessible components, allowing founders to focus engineering efforts on unique features rather than reinventing solved problems like keyboard navigation in dropdowns.