Search for a property in an Indian city using a plain-English description instead of filters.
Compare two shortlisted properties side by side with an AI-written verdict.
Check a property's nearby schools, hospitals, and metro access before deciding.
| tushar80rt/estate | 0xustaz/streamgate | a-bissell/unleash-lite | |
|---|---|---|---|
| Stars | 1 | 1 | 1 |
| Language | Python | Python | Python |
| Setup difficulty | hard | hard | hard |
| Complexity | 4/5 | 4/5 | 4/5 |
| Audience | developer | developer | researcher |
Figures from each repo's GitHub metadata at analysis time.
Requires Groq, ScrapeGraphAI, and NVIDIA API keys plus a running MongoDB instance, Google Maps is optional.
EstateGPT is an AI powered real estate search assistant built specifically for the Indian property market. Instead of scrolling through listing sites and filtering by dropdown menus, you type a plain English request such as a three bedroom flat in Bangalore under a certain budget with metro access and a good school nearby, and the system returns a ranked, explained shortlist of matching properties. Behind the scenes it pulls listings live from popular Indian real estate portals such as MagicBricks and 99acres, then cleans and stores them in a MongoDB database. It breaks each listing into chunks of text, converts them into embeddings, numerical representations that capture meaning, using NVIDIA's AI models, and stores those as well. When you search, it retrieves the closest matching listings, reranks them using another NVIDIA model to put the best matches first, and adds neighborhood details such as nearby schools, hospitals, and metro stations pulled from Google Maps. A large language model from Groq then writes an explanation of why each property was chosen and answers follow up questions. Beyond basic search, the tool can also compare two properties side by side and give a verdict on which is the better choice, calculate a rough investment and travel suitability score for a listing, and keep a history of your past searches saved to the database. The app runs through a Streamlit web interface, a Python framework for quickly building interactive apps, and requires several outside services to work: a Groq API key for the language model, a ScrapeGraphAI key for scraping listings, an NVIDIA API key for embeddings and reranking, a running MongoDB database, and an optional Google Maps API key for the neighborhood features. Setup involves cloning the repository, installing Python dependencies, filling in these API keys in a configuration file or the app's sidebar, starting MongoDB, and running the Streamlit app locally. This is a small project with 1 star, written in Python, and best suited for developers interested in building AI search tools or anyone specifically searching for property listings in India who is comfortable running a local Python application.
EstateGPT turns a plain-English property request into a ranked, AI-explained shortlist of real estate listings scraped from Indian property portals.
Mainly Python. The stack also includes Python, Streamlit, MongoDB.
Setup difficulty is rated hard, with roughly 1h+ to a first successful run.
Mainly developer.
This repo across BitVibe Labs
double-check against the repo, no cap.