Inspiration
Managing rental properties often means switching between long tenant conversations, AI chatbots, documents, and notice templates just to complete simple daily tasks. Important information gets buried in conversations, and writing professional notices repeatedly becomes time-consuming.
I wanted to build a single workspace where landlords could organize conversations, simulate tenant interactions before responding, and generate professional notice drafts without constantly switching between multiple tools.
What it does
Homie Desk is an AI-powered property management assistant that helps landlords manage tenant communication more efficiently.
Its core features include:
Conversation Highlights – Automatically organizes long conversations into meaningful highlights so landlords can quickly understand the discussion.
Tenant Simulator – Simulates realistic tenant interactions, helping landlords think through responses before communicating with tenants.
Notice Draft Generator – Generates professional notice drafts using available property and tenant information, providing a strong starting point that landlords can review and edit.
How I built it
The application was built using Next.js, React, TypeScript, and Tailwind CSS.
The AI features are powered by the Google Gemini API.
During development, OpenAI Codex and GPT-5.6 were used to accelerate implementation, debug issues, improve the interface, and refine the project documentation.
Challenges
One of the biggest challenges was integrating AI features while keeping the user experience simple and responsive.
Another challenge was debugging deployment and API integration issues, ensuring the application behaved consistently both locally and on Vercel.
Accomplishments
- Successfully built a complete AI-assisted property management application.
- Designed a clean and responsive interface.
- Integrated AI-powered conversation organization, tenant simulation, and notice generation into one workflow.
- Deployed the application publicly with production-ready documentation.
What I learned
This project taught me how AI can be integrated into practical workflows instead of simply acting as a chatbot.
I also gained experience using AI-assisted development tools like OpenAI Codex and GPT-5.6 alongside modern web technologies to rapidly prototype, debug, and refine a production-ready application.
What's next
Future improvements include:
- Multi-user authentication
- Persistent conversation history
- Calendar and email integration
- Property management dashboard analytics
- Multi-property collaboration for property managers
Built With
- api
- codex
- css
- gemini
- gpt-5.6
- next.js
- openai
- react
- tailwind
- typescript
Log in or sign up for Devpost to join the conversation.