Lofty AI (YC S19) – Real estate investment with alternative data

Hi Hacker News Community,

My name is Jerry, and I’m one co-founders for Lofty AI (https://www.lofty.ai/). We use machine learning to help identify homes where values are likely to appreciate, and we help home buyers buy them. People can partner up with us to buy a recommended property. If they do, we are willing to cover any potential losses on the property. In exchange, the buyer agrees to share some of the future profit on the home with us. The agreement lasts 3 years.

Before starting this company, my co-founder and I had tried to invest in homes. However, we quickly got tired of realtors telling us to make offers based on very little data. We wanted to figure out a way to buy affordable homes that had the highest growth potential via a data driven approach. We realized there was a wealth of new alternative data out there, which could be used to predict both neighborhood growth and individual property growth. This alternative data we envisioned ranged from the growth in the number of postings on social media about a specific dog breed, to the number of restaurants in an area serving a specific type of trendy food, to the average wait time for ride sharing apps, and the average maximum temperature an area can experience.

Our tech involves running clustering to identify trends and keywords from text based data (e.g.: social media photo tags, business reviews) that are associated with different categories of neighborhoods (for example: rich/suburban/static, middle-class/urban/growing). We then take these insights and feed them into a larger model with historical home prices, house level features, and an array of other numeric features (e.g. ride sharing wait times, new businesses) that predicts future home price on both an individual property and neighborhood level. With this trained model we can then predict future home prices based on these alternative data sources (as well a few traditional data sources). As we ingest more data going forward we are constantly retraining and reoptimizing our models. Along with successful backtesting we have been tracking our predictions to validate our models in production and have found that properties we had identified 12 months ago have beaten the market in appreciation by an average of 14 points (yay!).

Most young working professionals want to live in or near large metropolitan cities for the lifestyle and better jobs market. This has contributed to extremely high home prices for places like the bay area and many young professionals end up paying rent that is on par with a mortgage payment. However, instead of building equity in their own future through an investment, they are simply making their landlords richer.

We want to change this by giving people another option. They can now invest in a home and their capital can be protected should the investment flop. The trade off is that these homes tend to be located in areas not “currently” deemed to be a desirable neighborhood. In essence, we want to help inexperienced home buyers make smarter decisions, and we are willing to risk our own capital for that. In the event of a downturn in the market we are hedging our exposure by buying deep out of the money options that track the real estate market. These hedges are also attached to each individual contract so even if we were to go out of business before the maturation of the agreement or before a downturn in the market your downside protection would still be alive and well! As a result, anything that’s above a 20% decline across the portfolio will be covered by the hedging instruments, so we only need to be able to guarantee the range between 0 to -20% using our own capital. To make sure we can abide by the guarantee, we know exactly how many contracts we can enter into, and we will not go above that threshold until we obtain more funding.

Sign up with us to receive a list of recommended properties that our models think will appreciate over the next 3 years. Make an offer on the property you like the most using any method you’d like. If you don’t have an agent you work with, we can recommend you one along with helping you get a mortgage. After you make an offer on a home, you enter into a contract with us. We agree to cover losses over the next 3 years and in exchange, you share some of the future upside with us.

Let us know if you have any questions or insights, and I’ll be happy to respond! Feel free to directly reach out to me at [email protected] as well. We’d love to hear your feedback and suggestions!



Get Top 5 Posts of the Week



best of all time best of today best of yesterday best of this week best of this month best of last month best of this year best of 2023 best of 2022 yc w24 yc s23 yc w23 yc s22 yc w22 yc s21 yc w21 yc s20 yc w20 yc s19 yc w19 yc s18 yc w18 yc all-time 3d algorithms animation android [ai] artificial-intelligence api augmented-reality big data bitcoin blockchain book bootstrap bot css c chart chess chrome extension cli command line compiler crypto covid-19 cryptography data deep learning elexir ether excel framework game git go html ios iphone java js javascript jobs kubernetes learn linux lisp mac machine-learning most successful neural net nft node optimisation parser performance privacy python raspberry pi react retro review my ruby rust saas scraper security sql tensor flow terminal travel virtual reality visualisation vue windows web3 young talents


andrey azimov by Andrey Azimov