A comprehensive system for submitting and managing maintenance work orders with AI-powered features for transcription, location detection, and work order generation.

This project provides a platform for users to submit maintenance work orders through a user-friendly interface. The system leverages AI capabilities to enhance the work order creation process:

• Audio Transcription: Convert spoken descriptions into text using OpenAI's Whisper model
• Image Analysis: Process images of maintenance issues to include in work orders
• Location Detection: Automatically identify and match location references in work order descriptions
• Smart Work Order Generation: Create professionally formatted work orders from user inputs

• Authentication System: Secure user authentication and authorization
• Work Order Management: Submit, track, and manage maintenance requests
• Wand Devices: Support for physical maintenance "wands" that can submit work orders from the field
• Location Database: Comprehensive database of buildings and locations
• AI Integration: OpenAI-powered features for transcription and content generation

• /backend: Server-side code
  • /db: Database schemas and queries
  • /routes: API routes and handlers
  • /workorders: Work order generation logic
• /frontend: Client-side code
  • /components: React components
  • /pages: Page components
• /shared: Shared code between frontend and backend
• /pi: Code for Raspberry Pi devices (wands)

• Deno runtime
• OpenAI API key (set as environment variable)

• Initialize node modules: deno cache --node-modules-dir .
• Generate database migrations: deno run -A --node-modules-dir npm:drizzle-kit generate

• Enroll devices
• Implement vector similarity search: https://orm.drizzle.team/docs/guides/vector-similarity-search