Inspiration

Renting can feel extremely overwhelming, as you have to look through endless listings, send repetitive emails, and navigate through confusing platforms. We wanted to make the process faster, smarter, and less stressful as this is a problem very close to home for us as we are also looking for off-campus housing next year.

What it does

Rent AI uses Fetch.ai agents and Gemini to find the best rental properties based on user needs/preferences. Then, with AgentMail, it automates landlord communications to schedule viewings, request details, and streamline the leasing process.

How we built it

We collected data on rental properties from many different places to gather listings, used Gemini for parsing the data and doing property analysis, and finally we used Fetch.ai agents to create a chatbot so you could easily find rental properties based on your needs. Additionally, we also used the AgentMail API to automate communication and email conversations between the landlords and the consumers.

Challenges we ran into

Some challenges we ran into were handling large property datasets efficiently, integrating multiple APIS smoothly, automating the email system to keep them personalized and send without running the file, and getting the Fetch.ai agent to connect with the chatbot.

Accomplishments that we're proud of

We were very proud of building a successful end-to-end agent pipeline from property search to landlord communication. Additionally, we were also proud of automating repetitive renting tasks with Fetch.ai and AgentMail and how we designed an expert chatbot with platform-specific knowledge.

What we learned

We learned how to combine LLMs with agent frameworks for different practical tasks. Additionally, we learned the importance of clean data for property analysis. Lastly, we learned how to automate email conversations and make them sound natural and professional using AgentMail.

What's next for Rent AI

Something we are looking forward to do with Rent AI is expanding the API coverage so we can list and show properties from even more places. We also want to show even more specific details of each property and make everything even more tailored to the user's experience. We would also like to implement a SMS/voice communication support for our service. Lastly, deploying this project as a full-stack app for real renters to use would make this service even more accessible and efficient to use for everyone.

Built With

Share this project:

Updates