‘The answer to everything is always data’

This interview has been edited for length and clarity.

Sarah Wheeler: How do you determine on which technology you should concentrate on customers? What does that process look like?

Joe Trapani: Thinking from a view of 10,000 feet: everyone wants everything faster, cheaper, better. So everything we build must make us faster, cheaper, better.

My philosophy is very bad that the days of the technical components that are in a dark cupboard, where nobody speaks directly with them, are over. I have more conversations with CEOs than CTOs. We are with our customers and really dig in the problem room, understand what their primary problems, challenges and bottlenecks are and then make the solution.

A few years ago we started with what I call the blueprint strategy. Because Moder is within a number of different customers within all our verticals that we serve, we hear many of the same, common problems and we are not a platform company today. So what we have developed were blueprints that can easily be entered, can easily be configured in your area and solve a common problem. We now have a catalog of solutions that are really easy to implement with customers.

SW: Your team has expertise in insurance, banking and financial services. How does this broad expertise go to your mortgage customers?

JT: My biggest ‘aha’ that came in mortgage was that this is the only industry where knowing the company is more important than knowing the technology. So I think we have that background, especially in the advisory and outsourcing space in which we are in which we find ourselves, is invaluable. And I have never been a seller – I have always been there on the client side and have always done the build. So I have often seen suppliers come in and pretend to understand the space, but they really don’t do that, or they only understand at surface level.

We are so hyper -oriented on this vertical that when you come in and start talking to us, whether it is our activities or our technical side, you talk to experts. You talk to people who have been there, have seen it. We don’t start all over again. We do not learn what a loose is on the day we come. We also have a huge stable of not only operational expertise, but also technical expertise. So we can immediately give a lift manually and then gradually phase technology for our customers. It is something that we see enormous value in doing.

SW: What do you do with Gen ai that impact is?

JT: I am close to technology long enough to know that there is always this emerging ‘next big thing’ and today is that Gen Ai. In the beginning my greatest care with Gen AI was safety and privacy and compliance. We are in a very conforming industry, so one of the things on which we immediately concentrated was Privé -LMS. One of our blueprints is a private -llm that can be used in the surroundings of our customer. We do not want to have customer data. Every time someone owns your data, this is a risk. So we immediately developed a private -llm that can be dropped in the vicinity of a customer.

Now your agents and employees can have access to the typical Chat GPT functionality that we are all used to on the first day and that we love, but it is private. You know that the questions, answers and data that are uploaded or downloaded do not go over the internet. It all happens in the environment of people, and it is safe.

Then we adjust those LLMs and make them expandable and really the target. Our blueprint chatbot, which is MODER AI, can adjust it to specific use cases by making adapted knowledge bases for it. So you have your general and say that we use the scenario agency, or the conforming guidelines on which we can quickly train that model. And so users now receive targeted responses that quote sources.

We have also added things such as general document repos on which we can train that, so we have customers who use our chatbot in document repositories so that their brokers or even their employees can ask questions and receive tailor -made answers based on the documents that send their business forward.

This is a very risk-intolerant company, so we are not going to trust. We start with these small, targeted copilots or agents who can help. And now we go to the space of process automation with a person in the loop. Directing the right processes and introducing AI safely is one of our primary focus.

SW: How do you decide when there should be a person in the loop?

JT: The answer to everything is always data. When we use AI, we get a reliability level back. Over time, the data tell us when we can safely fully automate. There are processes that I think probably always require a person in the loop – for example, the final insurance decision. But there is a huge difference between the traditional process and AI that have a decision ready and that present information to an agent for their assessment. It is still an important increase in your productivity if you now assess what that AI can do.

And if you look at the rise of agentic AI, that is where that trend is going. This [AI] Agent is sitting there and it takes several steps workflows and automates that and then asks people: “Hey, this is what I think I’m going to do. What do you think?” And I think that is feasible in the very near future.

SW: What distinguishes moder?

JT: One of the things that I really think are the best is our customer obsession. We are dedicated, end -To -end, to our customers. We constantly want to push things out when we work with our customers on the technical side – it’s every two weeks. We want to be the partners we wanted. We do internal development in the same way – the technical side sits down with things and understand the business problem before we start building things. When people work with us, they really have the feeling that we are sitting with you as a partner and digging in understanding what you need.