EstateEdge AI

Smarter Real Estate Decisions — Personalized Investment Advice & Instant Property Price Estimates

Comment

🏗️ Project Story

🚀 About the Project

We built Estate Edge AI — a web-based real estate advisor that simplifies property investment and price estimation using the power of AI.

The inspiration came from a simple question:
“What if anyone could get personalized real estate advice and price insights instantly, without needing an agent?”

We wanted to empower users to make smarter investment decisions and estimate home prices with confidence — using just a few inputs.

The project has two main features:

Feature 1: AI Investment Planner

The first feature helps users understand how and where to invest based on their risk level, goals, and budget.

We designed a form with dropdowns and fields for:

  • Investment style (e.g., all at once, monthly)
  • Goal (growth, income, etc.)
  • Budget, duration, and risk level

Once submitted, the backend uses custom-built AI agents to:

  1. Summarize the user’s profile
  2. Generate two real estate investment strategies
  3. Predict the ROI over 1, 3, and 5 years

All of this is wrapped into a beautifully styled HTML report. We used GPT-4 to reason through the investment options, simulate returns, and format the output in a clean, readable way. The result? A report that feels like it came from a personal financial advisor — instantly.

This feature taught us how to:

  • Orchestrate multiple AI agents in a logical flow
  • Prompt GPT to produce clean HTML with no markdown
  • Style dynamic AI content in a readable and engaging way

Feature 2: AI Property Price Estimator

This part started out ambitious. We wanted to fetch live housing prices from Zillow or MLS. But here’s the catch — no free, reliable API was accessible for us within the hackathon time frame.

After hitting that wall hard, we decided to pivot.

Instead of live APIs, we:

  • Downloaded a real real-estate dataset for a U.S. city
  • It contained 20+ property features like price, location, square footage, bedrooms, etc.
  • We cleaned the data using pandas and created a simplified, useful dataset

Now, when a user enters property details (like bedrooms, bathrooms, region, etc.), our AI Price Estimator Agent compares it against the cleaned dataset and estimates a fair market price. GPT acts as the analyst — reasoning over the data and giving back an explanation + estimated value in plain English.

💡 This was our biggest challenge:
Going from "fetch real-time API data" to “simulate a live pricing experience” using pre-downloaded data — while keeping it useful and realistic.

🛠️ Tech Stack

  • FastAPI (Backend)
  • Jinja2 Templates (Dynamic HTML)
  • OpenAI GPT-4 Turbo (AI Reasoning)
  • Bootstrap 5 + Custom CSS (Frontend UI)
  • Pandas + CSV (Local data processing)

Built With

  • bootstrap-5
  • fastapi
  • html/css
  • jinja
  • nominatim-api-(openstreetmap)
  • openai-gpt-4-turbo
  • pandas
  • python
Share this project:

Updates