Stratus 2025 AI Hackathon

This is my project for the Stratus 2025 AI Hackathon, A-EYE - experimental feature playground for testing AI with images.

Project Installation

• clone repository
• run pnpm install to install dependencies
• Generate a Google Gemini API key and store it as GEMINI_API_KEY in .env
• run pnpm dev to launch project

Project Features

• Part Identificatation
  • Able to identify wide variety of objects without any additional training.
  • Training with real world data may improve results further.
  • Able to look up dimensions on known objects, but provides estimations for unknown objects.
• Document Scanning
  • Able to transcribe text from physical objects without any extra libraries for OCR.
  • Response can be returned in markdown, to give more formatting to scanned documents.
• Diagram Generation
  • Able to generate response based on the content of a photo.

Project Information

• Goal: Experiment utilizing AI to extract and analyze information from images.
• AI Model: Google Gemini
• Total Cost: Google Gemini Free Tier - $0.00
• Frameworks used: Astro, Vue
• Hosting: Netlify
• Created by: Ryan Trimble