GoSeek: A Mobile Scavenger Hunt Powered by On-Device AI

Developer 0shuvo0 has introduced GoSeek, an innovative mobile scavenger hunt designed to encourage outdoor activity through gamified riddles. Unlike typical apps that command users to go outside, GoSeek provides cryptic clues for everyday objects, requiring users to find and photograph them in real-time. The project leverages a hybrid AI architecture: Gemma 3 4B, running locally via Ollama, generates creative riddles, while the lightweight SmolVLM-500M vision model runs directly in the user's browser via Transformers.js to verify image matches. By processing photos entirely on-device, the app ensures user privacy and eliminates the need for server-side image storage. Built using Google AI Studio, React, and TypeScript, the project demonstrates the potential of open-weight models to create privacy-focused, cost-effective, and interactive experiences. The app is currently available for mobile users, turning urban exploration into a competitive game of discovery.
This is a summary. Read the full article at the original source:
Dev.toRelated stories
The author shares an unusual experiment in automating code generation using artificial intelligence. Instead of using a single tool, they created a sy…
While Anthropic’s Claude is strictly a text-based model that cannot directly generate pixels, it remains a powerful tool for visual workflows. In this…
As part of the Hacktoberfest Open-Source AI Challenge, developer Arnulfo explores a practical application for local large language models. The project…



