HomeReady

• 

  main_page

• 

  main_chatbot

• 

  crime_rate

• 

  school

• 

  house_mortgage1

• 

  house_mortgage2

• 

  house_mortgage3

• 

  compare_1

• 

  compare_2

Inspiration

Buying a home today is overwhelming. People jump between Zillow, school rating sites, crime maps, and Google Maps just to understand one property. We wanted to reduce this “tab overload” and answer a simple but critical question: “Can I actually afford this home?”

What it does

HomeReady is an all-in-one AI-powered platform for homebuyers. It combines location data (crime, schools, traffic, etc.) into a single interactive map and adds an AI assistant that helps users:

Discover better neighborhoods Understand total monthly costs Check mortgage readiness based on income and credit

Instead of guessing, users can make decisions with real data and clear financial insights.

How we built it

We built an interactive map using Mapbox to visualize multiple public datasets as layers. On top of that, we integrated an AI assistant using RAG + function calling.

Frontend: React + Mapbox for dynamic map visualization Backend: Node/Express with API integrations AI Layer: RAG pipeline + mortgage prediction model (HMDA-based) Financial logic: Custom mortgage calculations (PITI, DTI, affordability)

The system connects user inputs, map context, and financial models in real time.

Challenges we ran into

One of the biggest challenges was combining different types of data into a single experience. Location data, financial calculations, and AI responses all had different formats and latency.

We also realized that “approval probability” alone is not useful. Users care more about monthly cost and affordability, so we had to rethink our core feature mid-project.

Accomplishments that we're proud of

We’re proud that we turned a fragmented, stressful process into a single intuitive workflow.

Unified multiple datasets into one map Built a working AI assistant that understands user intent Shifted from a “toy predictor” to a real decision-making tool Created a product that has clear B2B value through high-quality leads

What we learned

We learned that solving the right problem matters more than adding features. At first, we focused on prediction, but users actually needed clarity and confidence.

We also learned how powerful it is to combine AI with structured financial logic instead of relying only on ML outputs.

What's next for HomeReady

Next, we want to improve accuracy and real-world usability:

Integrate real lender APIs for live rates Improve closing cost and “cash to close” estimation Add personalization (user profiles, saved searches) Build a lender handoff system to convert users into real applications

Our goal is to make HomeReady not just a tool, but a bridge between homebuyers and lenders.

Built With

• ai
• express.js
• machine-learning
• mapbox
• python
• rag
• react
• supabase
• typescript