Payhawk CEO Hristo Borisov explains why the company’s new Financial Controller Agent is built to cut routine complexity for CFOs while maintaining oversight. As the first in its AI Office of the CFO suite, the agent marks a strategic pivot toward AI-first finance.
Hristo Borisov isn’t selling a vision of AI built on big promises. As co-founder and CEO of Payhawk, he’s focused on something narrower: how finance leaders can use automation to cut through routine complexity without giving up oversight.
“Finance leaders don’t need another assistant that parrots policy,” he says. “They need a system that understands context and gets things done, under their control.”
Payhawk’s new Financial Controller Agent, announced this quarter, was built for that purpose. It is the first agent in the company’s AI Office of the CFO suite, which will expand with three additional agents scheduled for release in September. Unlike generic chatbots, the Financial Controller Agent doesn’t just summarize documents or offer search results. It takes action. The system chases missing invoices, flags incomplete transactions, proposes corrections, and enforces company policies—all within parameters set by the finance team.
The intent is not to replace CFOs. It’s to reduce the manual friction that pulls them away from high-value work.
The core idea behind Payhawk’s product is that most finance departments are still bogged down by routine tasks that don’t match the expectations now placed on their leaders.
“CFOs are expected to guide capital allocation and identify market opportunities,” Borisov says. “But they’re still pulled into the weeds by legacy tools.”
That tension (between strategic demands and operational bottlenecks) is where Borisov sees the clearest role for AI. But only if it’s embedded in the system, not bolted on after the fact.
Payhawk’s agent isn’t trained on general web content or corporate jargon. It’s structured around workflows: procedural, rules-based, and able to interact with the actual mechanics of spend management.
“This is not another experiment,” he says. “It’s a working layer of intelligence that lives inside the core workflow.”
Borisov is direct about the language problem AI vendors have created. “AI agent” is now so overused that it’s nearly meaningless. Most systems described this way, he says, don’t do more than surface information in a chat window.
“There’s a difference between something that can respond to a prompt and something that knows what to do next,” he says.
By contrast, the Financial Controller Agent operates continuously in the background. It learns from user behavior, highlights risk, suggests categorization changes, and escalates issues that fall outside policy. It does all of this without requiring a prompt, while keeping the finance team in control.
Critically, it doesn’t make silent changes. All actions are documented, auditable, and reversible.
“There is no AI without auditability,” Borisov says. “We’ve built this for finance, not for general productivity.”
Inside Payhawk, the launch of the agent isn’t being treated as a feature release. It marks a shift in how the company defines its future.
“We are now an AI-first company,” Borisov says, not as a branding exercise, but as a statement about where the platform is headed.
That’s not an uncommon claim in 2025, but Borisov’s framing is more technical than aspirational. He’s not trying to wrap a generative tool around a dated architecture. Instead, he’s betting on domain-specific systems, ones that reflect the structure and constraints of modern finance.
The market context is shifting as well. Incumbents are adding AI layers to ERP systems. New entrants are competing on interface rather than depth.
Borisov believes there’s a middle ground: a product that understands the logic of finance and is wired into execution, not just visibility.
“Speed, compliance, visibility – that’s the triangle finance teams live in,” he says. “If we can help maintain all three without them lifting a finger, that’s real value.”
Borisov avoids overpromising. He doesn’t claim AI will transform corporate finance overnight. But he’s confident that the right approach, quiet, targeted, and built into core systems, can eliminate much of the inefficiency that still clogs daily operations.
The Financial Controller Agent isn’t pitched as a chatbot. It’s not a virtual assistant. It’s a background system designed to handle the repetitive work of managing company spend, while keeping CFOs in full control.
As more vendors enter the AI space and “AI-first” becomes standard language, Payhawk’s approach is worth noting for its restraint. It may not get headlines for making bold claims, but it does offer a practical model for how automation can support decision-makers without replacing them.
In a category full of AI demos, Borisov is opting for something more grounded: a system that does less talking and more doing.