Inspiration
After discussing a couple ideas during the initial brainstorming phase, a team member that had recently moved to America shared his opinions about the current search to easily find a real estate agent that tailor suited to his needs. This sparked an idea to create a website that would provide basic information about real estate agents.
What it does
Instead of plainly listing out information about a real estate agent with hundreds of other listings, Realsy uses a 'Tinder-like' swipe left/right function to provide users with a simpler experience trying to initially find and contact agents, fast.
How we built it
The project was split into 3 distinct parts that connect together to establish a clean design. The front end is a website that was built using React.js, HTML, and CSS in Visual Studio Code. The website was made to look and represent the aesthetic of Realsy. The backend consists of a serverless database run on CockroachDB in conjunction with Prisma and Flask. This was written with Node.js and Python. We split up the tasks for the front-end and back-end which enabled us to have an efficient workflow. The third part of the project is the data collection, which is done through our scrapper in communication with the CockroachDB. The scrapper gets information from online websites and uploads them to the DB whenever necessary for the front-end application. In times in which it isn't necessary the front-end application will only call the database and not invoke the scrapping.
Challenges we ran into
While we had abstracted all our work into modularly, we encountered various issues especially with all the new tools we are working with. Our most common one is getting around with Git and the whole process of the using it in the "right way" as we have made many mistakes in our implementation of statements and processes which made up burn a lot of valuable time during the hackathon. We also had issues in understanding how to use the cockroachDB especially in communicating with the front end side of the application. Our website also had a couple issues due the variety of data that might be sent through web scrapping, as improper HTML formats from websites will provide us incorrect information, that have to be solved rather hardcoded.
Accomplishments that we're proud of
We are proud of using a modular approach in developing our web application, one in which each team member can contribute simultaneously and effectively towards our project. We also had robust code when creating our scrapper and communication programs with the database. It saved us plenty of time but also saved the run time complexity of pulling a single request of information regarding realtor agents in a specific city. We are also proud of using the CockroachDB to its full capacity by utilizing the elastic feature to run with a varying amount of information.
What we learned
- implementation of CockroachDB serverless
- connection between front-end and DB using flask and flask_orm
- how the react.js frameworks works in parallel with the front-end design
- UI design with CSS
- Data scrapping using the bs4 library
- sending and receiving data using SQL queries (to n from database)
What's next for Realsy
- We would like to update our website so that more realtors can be available to be viewed from a single GET request to the fron-end application.
- We had some minor bugs in creating alerts for users who are interested in a specific agent.
- Our plan is to host a full scaled server to host the web application
- We also want to create a in website communication platform for clients to connect with the real estate agents, instead of temporarily giving them just the contact information.
Log in or sign up for Devpost to join the conversation.