🛑 SafetySnap: AI-Powered OSHA Violation Detection

Spot OSHA violations before they become problems with AI-powered analysis for construction workers

🚧 The Problem

Construction sites are hazardous environments where safety violations can lead to serious injuries, fatalities, and costly penalties. Identifying OSHA violations early is crucial, but:

  • Safety managers can't be everywhere at once
  • Workers may not recognize potential hazards
  • Documentation and reporting are often cumbersome
  • Violations frequently go unnoticed until inspections or incidents

💡 Our Solution

SafetySnap empowers construction workers to be proactive safety advocates by allowing them to:

  1. Snap a Photo of potential hazards at their worksite
  2. Analyze Instantly using Gemini AI to identify OSHA violations
  3. Get Results with clear feedback and actionable recommendations

✨ Key Features

Feature Description
🔍 Real-time OSHA Violation Detection Instantly identify potential safety hazards and OSHA violations
📊 Detailed Compliance Reports Generate comprehensive documentation for safety records and reporting
📸 Photo Documentation Maintain a visual record of safety conditions for analysis and improvements
📈 Actionable Insights Receive practical steps to address safety concerns and maintain compliance
🛡️ Preventative Safety Measures Identify potential issues before they lead to accidents

🚀 How It Works

  1. Snap a Photo
    Take a picture of an area of concern at your construction site.

  2. AI Analysis
    The image is instantly analyzed using Gemini AI to detect potential OSHA violations.

  3. Get Results
    Receive detailed feedback highlighting any violations, along with recommended actions to improve safety and maintain compliance.

🛠️ Technology Stack

  • Frontend: Next.js
  • Backend: Supabase
  • AI Integration: Google Gemini API
  • Authentication: Supabase
  • Styling: Tailwind CSS

💻 Usage

  • Sign up for an account or log in
  • Upload a photo of a potential safety concern 1
  • Review the AI analysis for OSHA violations
  • Get recommended actions to address any identified issues
  • Save and share the results with your team

🔮 Flow Diagram

SafetySnap Flow Diagram

What We Learned!

• How to integrate advanced AI vision models (Google Gemini) to analyze construction site images for safety hazards

• Techniques for creating an intuitive user interface that requires minimal training for construction workers

• Methods for translating complex OSHA regulatory requirements into actionable safety recommendations

• Approaches to build offline-capable web applications for use in remote construction environments

• Strategies for balancing accuracy with speed when processing safety-critical image analysis

• The importance of user-centered design when creating tools for non-technical industries

• How to effectively present technical solutions to address real-world problems with significant human impact

👥 The Team

SafetySnap was created during a hackathon by Mahdi, Cole, and Noah.

📜 License

This project is licensed under the MIT License – see the LICENSE file for details.

Built With

  • next.js
  • shadcn
  • supabase
  • tailwind
Share this project:

Updates