A user successfully utilized Google’s Gemini AI to build a custom Android gardening application, demonstrating the potential and limitations of using generative AI for personal software development projects.

Key Points

• The author used Google’s AI Studio to generate a functional yard-management app through iterative text prompts.
• Key features included a plant-health diagnostic tool using image recognition and a task-tracking system for garden maintenance.
• The AI initially struggled with real-world logic, such as choosing legible color schemes and prioritizing live weather data over climate presets.
• The project required significant manual debugging and refinement to address functional issues like broken date pickers and incorrect task sorting.
• AI-provided advice correctly identified that landscape fabric and river rock were suffocating the user's plants, leading to successful garden rehabilitation.

Why it Matters

This experiment highlights the growing accessibility of "vibe-coding," where non-programmers can create functional software using natural language prompts. However, it also underscores that AI currently lacks a practical understanding of physical-world constraints, requiring human oversight to bridge the gap between generated code and real-world utility.