Fintech’s AI plays matchmaker between banks and new markets

The mobile intelligence company Flybits wants to play matchmaker between banks and third parties looking to share data.

The company announced this week it is building a marketplace designed to help banks, fintechs, telecommunications firms, retailers, ride-sharing services, grocery stores and many other firms to collaborate. Within the marketplace, firms can make their services available to each other, rather than forcing firms to work individually with each other.

Sharing their application programming interfaces with the marketplace will allow them to connect more easily to Flybits and take advantage of its AI capabilities, which banks could use to deliver third-party services to customers.

For example, while the marketplace is still in beta, the company imagines that grocery stores and retailers could surface products to customers based on their financial profile, said Hossein Rahnama, founder and CEO of Flybits. Bank customers, meanwhile, could set up bill pay or upgrade their telecomm account within their banking mobile app. Wealth management firms might give a customer an Uber Black service back to their home on a rainy day.

“There are AI capabilities that are useful to banks that have not been invented yet or commercialized,” Rahnama said. “We know for a fact that lots of new capabilities will be created in the coming years — models that will require less data and less training time.”

Data integration is the core of the company’s business. It’s also one of the top pressing needs for financial institutions, said Ron Shevlin, director of research at Cornerstone Advisors. This is in part why Flybits attracted $35 million in a Series C from investors like Mastercard, Citi Ventures and Westpac’s venture capital arm, Reinventure, the company announced on Tuesday. The round was led by Point72 Ventures with participation also from existing investors Portag3 ventures, TD Bank and Information Venture Partners.

AB - Flybits (2).jpeg

Flybits context engine processes the data that banks feed to it — including information from the mainframe, CRM systems, sensors, segmentation services and databases — and can feed that back into the bank’s machine learning models in real time.

The company’s “inference router” allows different machine learning capabilities to be employed and routed to the relevant machine learning node. It then determines the profile of the customer and the type of machine learning algorithm appropriate for them — so that an algorithm created for high-net-worth individuals is not applied to a college student with no credit.

The ability to orchestrate a bank’s AI capabilities was part of what makes Flybits attractive as an investment opportunity.

“There are a lot of built-in capabilities that already exist at banks,” said Ramneek Gupta, managing director and co-head of venture investing at Citi Ventures. “They’ve already gone through all the compliance pieces and gotten them deployed.”

Flybits' asset management clients are interested in developing capabilities that allow customers of a certain asset portfolio and location to request a financial adviser to meet them in the same way that they might hail an Uber.

While options like these and the ability to offer customers insights from retailers and other third parties might seem like a way for consumers to engage more with the banking app, it’s also asking customers to change their normal behavior, said Shevlin.

“Creating an Uber-like service might seem like a more sexy service,” he said. “But it’s not how consumers typically do things like shop for mortgages and shop for groceries. Plus, a lot of retail stores and grocery stores are creating their own mobile apps to interact with their customer base.”

With this in mind, the fintech is putting part of its funds towards creating a more simple design interface for its clients.

“Very similar to what happened in early internet days where building a website was first the job of a software engineer and a developer and then you developed tools such as Wordpress and Dreamweaver,” Rahnama said. “Creative people could now build great web applications and web 2.0 tools. We have now built a tool that does the same thing for banks in the area of AI and data science.”

For reprint and licensing requests for this article, click here.
Mobile banking Mobile technology Platforms Artificial intelligence BankAI Conference
MORE FROM AMERICAN BANKER