Editor-in-chief Sarah Wheeler sits down Sublime Chief Technology Officer Henry Li talks about the boundaries of artificial intelligence and how it can reshape the company’s roadmap.
Sarah Wheeler: How is Lofty’s technology different?
Henry Lee: Lofty is a platform product – this has been our vision from the beginning. We want to be the operating system for our users and their key operators – for agents, real estate agents and brokers, no matter how big they are.
Platform can mean many different things. At Lofty, it’s an end-to-end experience – from consumer discovery on the IDX portal all the way to nurturing the customer relationship in the CRM and converting that relationship into a real deal. Completing this transaction generates various operational insights in the system to drive informed business decisions.
Platforms also mean that we are more than just ourselves. We have a very robust marketplace bundle for our product, which consists of two parts. One is, obviously, we have an open API system, so we allow partners to build their own applications on top of that system. We also allow them to use other partner services through our open API system, and our marketplace offers many different value-added services.
One thing that makes us a little different is that we have pretty tight control over the user experience. So for the important value-added services – marketing automation services, automated marketing campaign lead generation services – those services are essentially created by us, but we also have a fairly open system. , distributing their services through local experiences.
The other thing about being a platform is that when we design the product, we think the product should be universal – we’re not just building products for the real estate industry. We’re a real estate technology company, so not only real estate agents can use the platform, but adjacent businesses like mortgage brokers and property management companies. We’re not building something like Salesforce, where you have to hire two Salesforce engineers to implement the entire system. In order to run a business on the platform, we need to do minimal setup.
The last thing is we’re a scalable platform, which means we’re able to provide an experience from a single user all the way up to an enterprise user with 10,000 users under that account.
SW: At the recent AI Summit, we divided AI into three levels: automation, traditional AI, and next-generation AI. How do you leverage artificial intelligence on these three levels?
HL: Of course, automation has been one of our selling points from day one. If you think about the traditional definitions of artificial intelligence, we certainly have a lot of those, like predictive analytics. We score lead insights to help clients understand the quality of the scores, and we integrate data enrichment services with third parties. There are also many recommendation systems embedded in our system, such as list reminders.
The last part is gen AI: specifically, we’re talking about large language models (LLMs). I think the most important thing is to really understand the boundaries of technology. We were one of the first vendors to launch Artificial Intelligence Generated Content (AIGC) capabilities, helping our clients generate listing descriptions, ads, blogs, newsletters, emails, and more.
I think artificial intelligence is very exciting because we have to completely dismantle our roadmap, I think three times at this point, because technology and technology service providers are making huge advances. Because technology is evolving so quickly, we must evolve with it. When you talk about artificial intelligence capabilities, it’s never a complete feature. Building artificial intelligence capabilities is different from the old concept of building software capabilities because you don’t necessarily have to write code. You are adapting artificial intelligence to give artificial intelligence more capabilities. So it keeps evolving.
Let’s first look at what types of jobs our users are hiring for. They often hire ISA companies [Inside Sales Agent] or remote virtual assistant. So our main focus is on AI assistants. Additionally, we have a sizable call center that provides first-level support to customers, and one of our AI capabilities is a service bot that can help with any product-related issues. We have a lot of conversations with users about how to diagnose specific problems, so we use these files to train the AI assistant.
But this is really just the beginning. What’s exciting is that we’re gradually being able to replace some of the tasks they give us in the call center. This will not happen 100% of the time, but with this type of infrastructure in place we will be able to gradually replace and enable artificial intelligence systems and gradually provide real, human-like services to our users. That’s really the second phase that we’re thinking about.
The challenge we face and the interesting problem we want to solve is that we think software is the perfect working platform for agents to obtain data, and we want to provide various tools to help them complete certain tasks. So what we need is a scalable way to deploy new capabilities for AI agents and deploy them into our product. We’re not talking about two AI assistants—what if there were 100, 200, or 300 different types of AI assistants helping our users with different types of tasks?
I think the biggest challenge we face going forward is that we need to redesign our software so that our users can access these features in a very intuitive way. So we talked about some of the features that we’ve released, but it’s really just scratched the surface compared to what we’re going to release.
SW: How are you adjusting to the changes in the NAR Settlement Agreement?
H.L.: From a software vendor’s perspective, the core parts of the software won’t change much. We’ve changed some small, very basic things, but now we’re talking about building functionality that makes sense to users in this environment.
First, we must help buyer’s agents better demonstrate their value. Now there’s a completely different journey – from lead acquisition to marketing. It’s important for our users to engage with customers as quickly as possible, so it’s important that you provide an experience for buyers to quickly review the agreement and agree to use you as a buyer’s agent – that’s what we want to optimize. We can help our users increase the speed of reaching agreements.
SW: What keeps you up at night?
H.L.: Understand the boundaries of this technology. That does change, so you have to think about some basic questions. Can AI replace agents? Will artificial intelligence replace software? Our current answer is no.
As an agent, you are building a business, not executing a mission. This is maintaining customer relationships and rain. So is artificial intelligence powerful enough to run a business? No. Can artificial intelligence replace software? Software and artificial intelligence are not really in conflict. Software is where users store data to manage their business. Software is essentially a workbench. So that doesn’t keep me going anymore.
What’s keeping me up at night today is that we don’t really know what the future of this industry will look like or how fast technology is evolving. As I said before, we’ve had to dismantle the road map three times. As a product company, we are looking for new paradigms or frameworks for building products. Building software now is a lot different than it used to be, so we stay vigilant and very focused. Honestly, it can be overwhelming because we are bombarded with new information every day. I really don’t get much sleep!
It’s a stressful program, but it’s also very exciting because as the environment changes and technology changes, we have new opportunities. Making sure we don’t miss these opportunities is what keeps me up at night.
