Google Maps gets smarter: New AI tools let developers build custom interactive projects

Google is bringing a wave of AI-powered tools to Google Maps, giving developers and users new ways to build interactive and personalised map-based experiences. The company announced that its Gemini models now power these tools, including a “builder agent” and a new MCP server that connects AI assistants directly to Google Maps’ technical documentation.

The highlight is the builder agent;  an AI-driven tool that lets developers describe the kind of project they want to build in plain language and instantly generates working prototypes. For example, users can simply type, “create a Street View tour of Paris,” “show a live weather map for my region,” or “list pet-friendly hotels in Mumbai,” and the agent will produce the code to make it happen. Once the code is ready, it can be previewed, exported, or refined using Firebase Studio.

A complementary styling agent allows developers or brands to create maps with custom designs, themes, or colour palettes; ideal for embedding on websites or apps that require visual consistency.

Google is also introducing Grounding Lite, a new feature that allows developers to connect their own AI models to Maps data using the Model Context Protocol (MCP). This means AI assistants can now answer spatial questions like “How far is the nearest hospital?” and present answers visually through Contextual View, a low-code component capable of showing results in list, map, or 3D formats.