Editor in Chief Sarah Wheeler sat down with Lofty CTO Henry Li to talk about the boundaries of artificial intelligence and how it’s reshaping his company’s roadmap.

Sarah Wheeler: What differentiates Lofty’s technology?

Henry Li: Lofty is a platform product — that’s been our vision from day one. We want to be the operating system for our users and their essential operating staff — for agents, Realtors, and brokers, no matter how big they are.

Platform can mean a lot of different things. At Lofty, it’s an end-to-end experience — from consumer search on the IDX portal all the way to nurturing the client relationship in a CRM and converting that relationship into a real transaction. Completing that transaction generates all kinds of operating insight into in the system to drive smart business decisions.

Platform also means we are not just by ourselves. We have a pretty powerful marketplace bundle with our product which comes with two parts. One is, obviously, we have an open API system, so we allow our partners to build their own apps on top of the system. We also allow them to use other partner services through our open API system and our marketplace offers a lot of different value-added services.

One thing where we are a little bit different is we have pretty tight control on user experience. So for the important value-added services — marketing automation services, automated campaign lead generation service — those services are essentially created by us, but also we have a pretty open system We are working with select partners to distribute their service through a native experience on top of our marketplace.

The other thing about being a platform is that when we design the product, we think the product should be generic — we’re not just building a product for the real estate industry. We’re a proptech company, so it’s not only Realtors who are able to use the platform, it’s adjacent businesses like mortgage brokers and property management companies. We’re not trying to build something like Salesforce, where you have to hire two Salesforce engineers to implement the whole system. We would require very minimum set up to in order to run the business on top of the platform.

So the last thing is we are a scalable platform, meaning we are able to offer an experience from a single user all the way to a corporate user that has 10,000 users under that account.

SW: At our recent AI Summit we categorized AI into three levels: automation, traditional AI and gen AI. How are you leveraging AI in the context of those three levels?

HL: Automation, of course, is one of our selling points from day one. And if you think about the traditional definition of AI, we certainly have quite a bit of that, like predictive analysis. We score lead insights to help customer to understand the quality of the score and we have integration with third parties on data enrichment services. There are also a lot of recommendation systems embedded in our system, for example, the listing alert.

The last part is gen AI: specifically, we’re talking about large language models (LLM). I think the most important thing is to really to understand the boundary of the technology. We were one of the first vendors roll out AI generated content (AIGC) features to help our customers generate listing descriptions, advertisements, blogs, text messages, emails, you name it.

I think AI is very exciting because we’ve had to completely tear down our roadmap I think three times at this point, because the technology and the technology service providers are making tremendous progress. Because technology is evolving so fast, we have to evolve with the technology. And when you’re talking about AI features, it’s never a complete feature. Building AI features is different than the old concept of building software features because it’s not necessarily that you’re writing the code. You are tuning the AI to give the AI more capabilities. So it’s ever evolving.

We start by looking at what type of work our users hire out. They often hire an ISA company [Inside Sales Agent] or a remote virtual assistant. So the main thing we’re focusing on is the AI assistant. Also, we have a pretty sizable call center that serves as tier one support for customers and one of our AI-enabled features is a service bot that helps with any product-related questions. We have a lot of of conversations with users on how to diagnose a particular problem so we’re using those documents to train an AI assistant.

But this is really just the beginning. What’s exciting is we are gradually able to replace some of the tasks that they have passed to our call centers. It’s not going to happen 100% but with this type of infrastructure, we’re able to replace and enable that AI system piece by piece, and gradually provide a real, human-like service to our users. This is really what we consider stage two.

The challenge for us, and the interesting problem for us to solve, is we think software is a perfect work bench for agents to have their data and we want to provide all kinds of tools to help them to complete certain tasks. So what we need is a scalable approach to deploy new capabilities for AI agents and deploy them to our product. We’re not talking about two AI assistants — what if it is 100, 200 or 300 different types of AI assistants that help our user handle different type of tasks?

I think, the biggest challenge for us in the future is that we’ll need to redesign our software to make those capabilities accessible to our users in a very intuitive way. So we talk about some of the features that we have already released, but compared to what we’re going to release, it’s really just scratching the surface.

SW: How are you adjusting for the changes that are part of NAR’s settlement agreement?

HL: From a software vendor perspective, it’s not going to change a whole a lot on the core piece of the software. There are some small, very basic things that we’ve already changed, but now we’re having discussions about building features that make sense for our users in this environment.

First of all, we’ll have to help buyer agents to present their value better. There’s now a whole different journey — from lead acquisition and from the marketing side. It’s important for our users to engage with the customer as soon as possible, so you will have to offer the experience for a buyer to quickly view the agreement and agree on using you as a buyer agent — this is what we want to optimize. We can help our users improve their speed to agreement.

SW: What keeps you up at night?

HL: Understanding the boundary of this technology. That’s really changed so there are some fundamental questions you have to think through. Is AI able to replace the agent? Is AI going to replace software? Our current answer is no for both of them.

As an agent you are building a business, not executing a task. It’s maintaining a customer relationship, it’s rain making. So is AI so powerful that it can run a business? No. Can AI replace software? Software is not really in conflict with AI. Software is where users store their data to manage their business. Software is essentially a work bench. So that doesn’t keep me up anymore.

What keeps me at night today is that we don’t really know what the industry is going to be like in the future and how fast technology is evolving. As I said earlier, we’ve already had to tear down our roadmap three times. As a product company, we’re looking at the new paradigms or framework for building a product. Building software right now is very different from the old days so we are staying vigilant and very focused. To be honest, it can be quite overwhelming because we’re receiving new information every day. I don’t really sleep a lot!

It’s stressful planning, but at the same time, it’s very exciting because as the environment changes and technology changes — we have new opportunities. Making sure we don’t miss those opportunities is what keeps me up at night.