This is my implementation of a real-estate AI agent for NYC!

Install the libraries in the requirements.txt file. Then open HomeMatch.ipynb and simply follow along. Here are the steps followed in the notebook.

Steps:

1- We're going to generate realistic apartment listings using an LLM. 
2- We're going to create a database of listings. The listings are also saved in a Listings.txt file
3- We're going to ask user's questions to capture their preferences for what kind of apartments they're hoping to find. 
4- We're going to summarize the user's questions and the answer to create a query vector that describes the user's preference. 
5- We're going to index our database using the query embedding and find the top 5 listing that match the user's preference. 
6- Finally, we're going to augment the description of the apartment with the user's preference and create personalized descriptions of the apartment.