Peach AI

Inspiration 💡

As a landlord or property manager, you probably know that finding the right tenant is essential to making money and keeping your properties in good condition. You also know that finding a good tenant often means sorting through tons of applications, knowing what details are important to look at, and being able to make a smart decision.

But that comes at a cost. Most Landlords and Property Managers admit that they spend more than 10-15 hours a week actively trying to find a qualified tenant. They’re constantly sending out endless email threads, sorting and reviewing tenant applications, and making endless phone calls with the goal to find someone who is kind, reliable, and trustworthy. And actually, good at paying rent on time.

But let’s assume you did all that leg work, now it’s time to assess, review, and qualitatively order the applicants. You spend countless hours thoroughly reviewing every application to try and pick out the ones that seem like they might be a good fit. But you don't have time to try and calculate everything yourself, and you also don't want to just guess. But the truth of the matter is, the decision to consider or dismiss an applicant is often made on the basis of ‘gut feeling’ instead of using data. This can lead to moving incompetent applicants further into the rental process while denying the tenants who are qualified, good candidates.

Before you accept your dream tenant, it's very likely that you've read something along these lines more times than you can count "I would like to see the apartment. Let me know when a good time for you is." It usually happens via email or text message, and before long, your phone is pinging so much that it's impossible to keep up with. Your time is an asset, and you need to be able to schedule showings quickly and efficiently so that your tenants can view the property as soon as possible, and in turn rent it.

While it may feel like there's nothing you can do about all this, we're here to tell you that It doesn't have to be that way and that there is something you can do about it!

What if there was an AI that could save you hours on end by calling all the applicants to gather their preferences and other critical data?

And what if that technology could tell you who’s a fit and who’s not, for your rental property, as soon as they applied?

What if that technology could even make all the other arrangements such as taking a glance at your calendar and scheduling the house viewings for you, so all you had to do was sit back and wait while it took care of everything else?

We all make bad or hasty decisions from time to time. As a landlord and property manager, don't let that choice be your tenants and that’s why we built Peach AI.

What is Peach AI 🤖 🍑

Peach AI is an intelligent personal assistant powered by AI for landlords and property managers. Peach helps landlords and property managers save time and money by automating their communication with potential tenants.

Our goal is to replace the manual work that goes into searching for, evaluating, and approving quality tenants with smart automation and machine learning. We want to take the stress out of finding new tenants and help landlords secure a high-quality tenant.

Peach AI has a wide range of advanced features and services, including intelligent screening and smart scheduling for house viewings with prospective tenants. Peach AI will call your preferred applicants on your behalf to gather information about their situation, then qualitatively order the applicants to help you choose the best candidate and even schedule a house viewing based on your future availability!

How to use Peach AI? 🤔

Landlord/Property Managers set up Peach AI.

a. User enters the ad details, including:

• Price • Location • Neighborhood amenities • Property description

b. Uploads pictures of their property c. Puts in relevant questions to ask potential tenants

Enable Peach AI

Now what?

When Inquiries come in via the platform, Peach AI takes over, calls all interested applicants, and asks them a series of questions to determine if they're a good fit for the place. Once the information is gathered, our algorithms do all the heavy lifting to give you a qualitative analysis of the data and provide you with a list of top applicants along with their answers. You’ll never have to take your chances or guess again!

Screenshots

How we built it ⚙️

Our beautiful and elegant mobile application was built using a cross-platform framework flutter using Dart programming language.

Additionally, we developed an AI that uses Twilio API to call all the tenants and ask them questions that the landlord has. The landlord would select which tenants to interview and AI would only call those tenants. We also developed another AI that would call selected tenants for a house viewing and schedule time and date for it. Our ai uses Natural Language Processing, Emotional Speech Synthesis, Natural Language Understanding, Voice cloning using Dasha SDK.

We also used a bit of NodeJS and JavaScript to set AI up as a server that would accept input parameters and run the model.

TechStack

Challenges we ran into 🧱

It was very difficult for us to learn about Machine Learning and AI and it was our very first time developing something at this scale. Developing a flutter application was also very difficult since no one in our team was good with graphic design. We managed to improve on our teamwork by actively discussing how we are planning to build it and how to make sure we make the best of our time.

Accomplishments that we're proud of ✨

we learned various tools such as Flutter, Dash.ai, Vscode, Twillio, Dart, and Javascript to create an AI app that allows you to filter out potential tenants being landlords. With this being our very first hackathon, we were pretty nervous competing against more experienced hackers however this was an extremely crucial learning moment where we learned the implementation of various software onto a usable application. We were able to observe our creation being brought into motion through what we had learned. What we believe we took from this hackathon was building our confidence in various skills to further go forward in our coding career.

What we learned 🙌

We as a team were able to develop an operational conversational AI bot within 36 hours. Our team is majorly comprised of beginners and we are proud of the fact that we utilized multiple APIs and open-ended languages such as Node Js to implement the project. We are proud of the fact that this bot can be utilized within any industry where there is a direct customer-to-service provider relationship. This bot will help save intense amounts of time on the user's end, hence solving a real-world issue.

What's next for Peach AI 🚀

There are repetitive tasks all around the world and AI can be used everywhere for this intention. We are planning to build similar models for Doctor/Patient, Interviewer/Potential Candidate, and many more!