Stability AI

Clear steps for tough times. Understand your tenant rights, build your defense, and protect your home.

Comment

Stability AI – Housing Support Platform

Inspiration

Millions of tenants face eviction every year without understanding their legal rights, court deadlines, or available support resources. Many people are overwhelmed, stressed, and unable to afford legal assistance when they need it most.

We wanted to create a tool that could provide immediate guidance, simplify complex housing information, and help vulnerable tenants take action before it is too late. This inspired us to build Stability AI, an AI-powered housing support platform designed to help tenants understand their situation, create defense plans, and connect with local resources.

What It Does

Stability AI is an anonymous, AI-assisted platform that helps tenants:

  • Understand their tenant rights
  • Generate personalized eviction defense plans
  • Track important legal deadlines
  • Assess eviction risk levels
  • Prepare for court appearances
  • Generate landlord communication letters
  • Discover nearby legal aid, shelters, rental assistance, and community resources

The platform transforms complex housing challenges into clear, actionable steps.

How We Built It

Frontend

  • Next.js 15
  • React 19
  • TypeScript
  • Tailwind CSS
  • Lucide React

Backend

  • Node.js
  • Express.js

Database

  • MongoDB Atlas
  • Mongoose
  • MongoDB Memory Server fallback for local development

AI Layer

  • Google Gemini 1.5 Flash

The application uses a multi-step intake process to collect housing, eviction, and financial information. This data is securely processed and sent to Gemini, which generates a structured action plan containing:

  • Urgency assessment
  • Rights summary
  • Required documents
  • Prioritized action steps

When AI services are unavailable, the platform automatically switches to a locally generated fallback plan to ensure uninterrupted access.

Challenges We Faced

One of the biggest challenges was generating reliable and structured AI responses. Large language models can sometimes return inconsistent output, so we designed a strict JSON schema and implemented validation and fallback mechanisms.

Another challenge was maintaining user privacy. Since housing issues are highly sensitive, we built the platform around anonymous sessions rather than requiring personal accounts.

We also had to balance simplicity with legal complexity. Housing laws vary significantly between jurisdictions, making it difficult to provide guidance that is both understandable and useful.

What We Learned

Through this project, we gained experience with:

  • Building full-stack applications using Next.js and Express
  • Integrating generative AI into real-world workflows
  • Designing resilient systems with fallback strategies
  • Structuring and validating AI-generated JSON data
  • Creating user experiences for high-stress situations
  • Working with MongoDB and scalable API architecture

Most importantly, we learned how technology can be used to improve access to information and support people facing difficult life circumstances.

Impact

Our goal is simple:

Knowledge is protection. Action is power. Housing stability starts with informed decisions.

Share this project:

Updates