HeatPumpBuddy was created by Mitchell Dodell, Rachel Plotnik, Sonali Anderson, and Daniel Hariyanto for the HeatSmartAlliance.

We were inspired by heat pump calculators, which allow people to put in information about their home and see how much different heating options would cost them on an annual basis.

However, the limitation to these calculators is that they don't actually encourage you to take action with the information that you are given. Our website empowers homeowners to learn more and provides them with the ability to take direct action by connecting them with coaches in their community who will work with them to implement a heat pump in their home.

What it does and how it helps the HeatSmartAlliance

Our application fulfills two needs: it gives the HSA an up-to-date and visually appealing web presence, while also being a resource hub for someone new to heat pumps to learn about why they are important and how they can be beneficial to them.

The HSA can connect a CSV file of their heat pump coaches and their information to our machine learning algorithm, so that way homeowners can be connected with coaches that best suit their personal needs.

What we each did

Sonali - conducted research on heat pumps and served as our environmental consultant

Daniel - worked on creating a nearest neighbor algorithm to match homeowners with a heat pump coach who has a similar home to them

Rachel - did the entire design for the website, produced the video, and focused on how we should market and design this for our end-users

Mitchell - designed and implemented Rachel's designs on a live SEO optimized website

How we built it

HeatPumpBuddy was built using the NextJS and React framework, along with a Flask machine learning API. We designed the application in Figma before bringing it to life through our development process. It is fully responsive on both mobile, tablet, and web so anyone can use this website at any time - whether they are researching options at home or on the bus to work.

The machine-learning algorithm allows users to enter information about their home and matches them with 3 heat coaches using a nearest-neighbor algorithm, so they can choose the coach who best suits their needs and preferences. The website is SEO optimized with Next, so someone looking up heat pumps would find our website and be able to learn more.

The website is hosted on Netlify for easy deployment, while the API is hosted on Heroku.

Challenges we ran into

The largest difficulty we had was designing a good website that was feasible to build within 24 hours. We had a lot of great design ideas, but it was very difficult to make something look beautiful and exactly how we wanted it within the span of DeisHacks.

Additionally, we were trying to figure out a way to actually calculate the cost of a heat pump versus alternative options directly on our website to provide the best user experience. Unfortunately, we were not able to devise an accurate algorithm, so we instead worked around that problem by linking to an external calculator.

Accomplishments that we're proud of

We are proud that we were able to create a fully responsive website that connects to a fully-working machine learning algorithm. Additionally, our website is stunning and has a professional look to it, and actually gives the end user a call to action, which is the most important part. We wanted to simplify the process of getting connected with the HSA, as the best way to convince someone to get a heat pump is by actually talking to a friend or advocate.

What we learned

This was the first time that we created a NextJS website that calls an externally hosted API.

This was also our first time working with SVGs to create beautiful backgrounds, which caused a lot of difficulties. Additionally, we wanted to make sure that our app worked on all browsers, and so we had to do a lot of work due to how certain web elements (SVGs, number inputs) interact with web browsers such as Safari, Chrome, and Internet Explorer.

Additionally, for half of our team, this was the first time even hearing about heat pumps. There was a lot of learning that we had to do to educate ourselves about our users' needs and why this was important.

What's next for HeatPumpBuddy

One thing we would like to do is provide HSA coaches with the ability to log in and see who they were matched with. Rather than the prospective owner reaching out, they can simply click a button that would say "I would like to match with this user" and it would connect to a backend database where the coach could log in, and see homeowners and their specified needs.

We would also have liked to have incorporated the heat pump calculator directly on the website but had some difficulty finding the correct algorithm that calculates the cost of various heating functions.

Built With

Share this project: