Few industries have had more money, talent, and technology poured into them than banking, and yet managing your finances still feels like a chore. For the people who run banks, it’s even worse. The work behind the scenes has only gotten more manual, complex, and expensive.
We believe that banks should run on auto-pilot. The customer operations, compliance, and back-office work that consume most of a bank's energy today should be handled by AI agents that are provably compliant at every step, freeing teams to focus on what matters most to their customers, and deliver it at a fraction of today's cost.
In pursuit of this vision, we’re announcing that we've increased our Series A to $26 million, led by new investors Octopus Ventures and CommerzVentures, with follow-on backing from Redpoint Ventures and Exceptional Capital. This funding will help us build the foundation for autonomous banking that drastically improves the experience of bank operators and the customers they serve.
Why we bet early on vertical AI
I started Gradient Labs in 2023, after joining Monzo as an early employee, helping build the bank from scratch. It was here that I met my co-founders, Neal Lathia and Danai Antoniou, and where we saw first-hand how all-consuming it is to operate a bank at scale. Are we screening signups correctly? Are we monitoring transactions in the correct way? Do we treat customers fairly and have the right controls in place for everything we do?
The reality is banks and fintechs spend most of their time and resources dealing with operational complexity. Even neo and challenger banks, built from scratch to do things differently, eventually reach a scale where the burden of operations overwhelms their ability to innovate.
Our conviction to solve these problems converged with the development of frontier AI models, and it was then that we realised two fundamental truths.
The first: technology could finally lift banks out from under the operational burden that has held the industry back for decades.
The second: general-purpose, horizontal AI wouldn't make a dent in it.
From these principles, we built Gradient Labs, and bet on vertical AI to tackle the most challenging problems in finance.
That bet has paid off. Over the last year revenue grew 900% and our agents now help over 32 million end users across some of the largest AI deployments in financial services. We’ve expanded our UK and European customer base of Wise, Zego, Monzo, Pockit (and many others) with standout US neobanks and fintechs like Current, Stash and Rho.
This round's transatlantic cap table, led by investors who have backed category-defining fintechs, is one of the strongest validations yet that vertical AI will be a crucial part of the future of financial infrastructure.
A suite of specialist agents for financial services
Now, we've taken vertical AI a step further: from a single agent to a suite of specialists, each built for one regulated job, to automate the complex, long-running workflows a general-purpose agent can't take on.
In production today, our Lending Agent automates the entire borrower lifecycle, from a missed payment to outbound collections calls, to an agreed repayment plan. Our Disputes Agent handles everything from intake to investigation to chargeback. A KYB Agent runs identity and document checks. These agents work as one system, sharing context and memory and passing work between each other where a customer's journey requires it. And they also work collaboratively with customer’s existing AI agents, in a true multi-agent fashion.
Our agents work, reliably and safely, on every channel a bank's customers use. We're one of the only companies running voice AI in production at scale in financial services. This is the hardest place to automate regulated work, and where we're furthest ahead.
Our success is only possible (from both a technical and regulatory standpoint) because of the domain specificity we've embedded into our specialist agents. Each agent includes the guardrails, compliance checks, and test scenarios for its domain, from FCA Consumer Duty to the EU AI Act. This is why so many organisations trust us to automate their long-running processes, and why we're doubling down even further on domain-specific AI agents for financial services.
The proof is in performance, and we stake our company on it
Importantly, it's not just scale or cost savings our customers are after, but a meaningful improvement in customer experience. In every single deployment across our customer base, customer satisfaction scores are higher than human teams, with some customers seeing scores up to 98%.
We're confident enough in our product to do something unusual in enterprise software: once we've scoped a use case, we guarantee the deployment. If we don't deliver what we said we would, customers get their money back. We share the risk because we're asking them to bet on us, and if we don't earn it, we don't deserve to be paid.
From co-pilot to auto-pilot
While we believe specialist agents will eventually run the entirety of a bank's manual operations on auto-pilot, the industry has mostly bet the other way for now, assuming AI is safer as a co-pilot. We've spent three years proving the opposite: that full autonomy is what delivers safer, more compliant, more delightful outcomes. Our CSAT and QS scores prove it.
Imagine applying for a mortgage in 2032. You apply at 11 pm and by the next morning, you wake up to an approved offer, one step away from moving into your dream home.
Behind the scenes, an AI agent verified your identity, traced the source of funds, screened for fraud, modelled affordability, and issued the offer. Every step was compliant, auditable, and explained in natural human language.
No one was pinged with a new task, you didn’t need to wait for days on end, nervous your application wouldn’t go through, and the bank spent that day building a better product instead of clearing tickets.
That's the world we want to live in, and the technology to do it already exists. What's missing is the operating system that ties it together, holds up to regulatory scrutiny, and earns the trust of the institutions that have to bet their licence on it. With this funding, we’re building it.
Come build with us
We're hiring people who want to rethink how financial services operates from first principles. If you've ever been frustrated by how slow, expensive, and impersonal banks are and you suspect (as we do) that it doesn't have to be this way, we should talk.
To our customers, our investors, and the Gradient Labs team: thank you. Three years in, we’ve built a product that delivers on its promise. It's also dwarfed by what's still ahead.
If you run customer operations at a financial services company and want to see how our agents handle your most complex cases, book a demo with us. We'd love to show you what’s possible.
Share post
Copy post link
Link copied