Dimitri Masin says he joined Monzo as roughly one of its first 30 employees.
“I've joined them quite early on, probably as one of the first 30 people, and was kind of lucky to bet on the right horse, to be honest.”
Person
Dimitri Masin is the CEO of Gradient Labs, and this page collects what he said on the Fintech One-On-One podcast about using AI in banking. He talked about why he believes fully autonomous AI agents beat co-pilots for customer operations, and about his earlier years at Monzo.
Masin says he joined Monzo as one of roughly its first 30 employees and stayed seven or eight years. He says Gradient Labs works with about 30 fintech and financial services customers. He also says its first full-scale deployment, at a large UK bank, left customers about 15 percent better off than the previous human team average.
1 episodes · 18 quoted claims
Dimitri Masin says he joined Monzo as roughly one of its first 30 employees.
“I've joined them quite early on, probably as one of the first 30 people, and was kind of lucky to bet on the right horse, to be honest.”
Dimitri Masin says he spent seven or eight years at Monzo while it grew from about 30 to about 4,000 people.
“And I spent there seven or eight years literally building the bank from 30 people up to 4 ,000 people, I think by the time I left.”
Dimitri Masin says the release of GPT-4 in early 2023 convinced him that automating bank customer operations had become possible.
“at the beginning of 2023, when GPT-4 came around, it became very, very clear to us that that part of the problem”
Dimitri Masin gives an example in which about 15 percent of KYC cases cannot be processed automatically and need deeper human investigation.
“there will be, let's say, 15 % of cases which cannot be straight through process because they need to be to look deeper at.”
Gradient Labs works with about 30 customers in fintech and financial services, according to Dimitri Masin.
“we have about 30 customers within FinTech and financial services space that we're working with.”
Dimitri Masin says a thorough deployment can save up to 70 percent of operational handling time on a process, while a minimal one might save about 20 percent.
“you can save maybe up to 70 % of handling time, operational handling time on a particular process.”
Gradient Labs says a proof of concept, where banks compare one or two vendors, typically takes about four weeks.
“So a bank would try one or two vendors and compare the performance that they can achieve. That POC phase is typically like four weeks”
Dimitri Masin says Monzo tried a co-pilot that suggested replies to customer support staff years before GPT-4 was released.
“And we actually tried to do it at Monzo years back, even before GPT-4 came around.”
Dimitri Masin says that when a co-pilot is correct about 90 percent of the time, staff tend to accept its suggestions without checking them.
“what tends to happen is that people start just accepting blindly what it suggests.”
Dimitri Masin says Gradient Labs has spent three years trying to prove that full autonomy is safer and higher quality than co-pilots.
“But now we essentially, for the last three years, we have spent the last three years proving that full autonomy is safer and higher quality than co-pilots.”
Dimitri Masin says none of Gradient Labs' customers have planned or carried out layoffs in the areas it automates.
“none of the companies we have worked with actually planned or laid off any people in that space, which is maybe like counterintuitive at first.”
Dimitri Masin says Gradient Labs had its first full-scale AI agent deployment at one of the largest regulated banks in the UK, starting around January 2025.
“it was already more than a year ago in January, we realized we have actually had the first full-scale AI agent deployment was one of the largest regulated banks in the UK.”
Dimitri Masin says the unnamed UK bank's customer experience under the AI agent was about 15 percent better than its previous human team average.
“I think it was like 15 % difference or better than the human team, the human average team in that particular bank previously.”
Dimitri Masin says that until about a year or so ago, larger banks typically told him they would build AI agents themselves.
“the typical conversation that I had with fintechs or banks probably one and a half years ago, one year ago, was like, especially tier one, tier two banks, we're definitely going to build it ourselves.”
Dimitri Masin says automating the easiest 50 percent of support queries yields only about 20 percent of the possible time savings.
“But the first, those 50 % of the queries that you're automating actually yield roughly 20 % of the time savings that you achieve through that.”
Dimitri Masin says even fast-moving tech customers take about a year to raise support automation from 50 percent to 80 percent.
“And it still takes a year to go from 50 % to 80 % where you really kind of see step changes.”
Dimitri Masin says US fintechs commonly run head-to-head bake-offs between vendors, whereas European firms typically pick one provider.
“In the US, I found that companies are a lot more performance-oriented. So essentially they do those bake-offs, essentially.”
Dimitri Masin says Gradient Labs has won every bake-off it has run against other vendors.
“We actually have won every single bake-off that we run against any other company.”