GeoSolvr

Project Overview

GeoSolvr is a multi-agent AI system that takes a GeoGuessr image URL and determines the location depicted.
Multiple specialized agents work together to analyze visual clues and reason about geography.

Features

• Multi-agent architecture with specialized reasoning agents.
• Analyzes visual cues like road signs, architecture, and vegetation.
• Accepts a standard image URL as input.
• Returns a predicted country and location with reasoning.

What I Learned

• How to build and coordinate a multi-agent AI pipeline.
• Prompt engineering for visual reasoning tasks.
• How to break a complex task into specialized sub-agents.
• Working with AI APIs and image inputs.

View the Code

The full source code is available on GitHub.