.png){kind=logo}
Rather than giving retailers another dashboard, Akaoglu wants AI to become the operating layer that helps enterprises decide, act, and learn.
Inside many modern companies, growth is treated like a collection of separate bets. The CRM team watches one set of numbers. Paid media watches another. Product teams measure behavior in their own language. Retention teams look for warning signs. Ozgur Akaoglu, co-founder of Hilbert, sees a deeper problem in that structure: activity is everywhere, but the system that connects it is often missing.
“Every company has growth teams,” Akaoglu says. “Almost no company has a true growth system.”
Hilbert was built around that idea. The company creates AI-powered growth intelligence for retailers and e-commerce brands, using customer data to identify high-value moments and coordinate AI-driven action across the business. Its platform is designed to help brands move from scattered signals to decisions that can be tested, measured, and improved over time.
Akaoglu is not interested in AI as another layer of commentary. He believes the next wave of enterprise AI will be judged by whether it can help companies make better decisions from a shared understanding of the customer.
“The opportunity is not just better intelligence,” he says. “It is a system that lets growth teams make decisions from the same understanding of what is happening.”
That point matters because growth inside large companies is rarely as coordinated as it appears from the outside. A brand may have strong teams, expensive software, and years of customer data, yet still rely on averages and intuition when deciding how to act. A customer may look promising in one system, fading in another, and invisible in a third. By the time the pattern is clear, the commercial window may already be closing.
Akaoglu calls that the move away from “faith-based growth.” Companies often assume that if each function improves its own slice, the business will grow in a compounding way. Hilbert’s argument is that this assumption breaks down when the pieces cannot learn together.
“Each team can optimize its own area and still miss what is happening across the customer relationship,” Akaoglu says. “Growth should not depend on everyone hoping the separate pieces add up.”
Hilbert’s answer is to build a unified model of how a business grows. That means looking beyond isolated campaigns or dashboards and connecting customer behavior, timing, channel, predicted value, and next action. The goal is not simply to identify a pattern. It is to move from prediction to coordinated execution before the window closes.
For retailers and e-commerce brands, the stakes are immediate. Customer attention is expensive. Acquisition costs can be unforgiving. Relationships weaken quietly before revenue loss becomes visible. A company may not know where it is losing traction until after it is already reacting.
“Most companies do not know which customers they are about to lose,” Akaoglu says. “By the time they realize it, the business is already reacting instead of deciding.”
Hilbert uses AI agents to manage that decision layer. The agents do not exist as a novelty or a demonstration. They are designed to work inside the brand’s real data environment, where customer behavior, value, timing, and channel choice have to come together. The point is not to give a team one more place to look. It is to help the business act with more precision.
That is important in a market crowded with AI tools that promise intelligence but still leave the hard work of execution to the client. Many businesses already have reports, models, segments, and dashboards. What they lack is a system that can connect those inputs, learn from outcomes, and keep improving the way the business grows.
“Dashboards can tell you where to look,” Akaoglu says. “A growth system should help you decide what to do.”
His view is shaped by years spent building predictive systems, first in operational environments and later in LLM-based product development. That background gave him a practical understanding of what AI can do, but also a sharp sense of what businesses actually need from it.
“The interesting part was never AI in isolation,” he says. “The interesting part was applying it to real business problems where the answer has to survive contact with the company’s data, teams, and constraints.”
That combination shaped Hilbert’s larger ambition. The company is not trying to sell AI as a single function or a narrow tool. Akaoglu sees it as part of a broader operating layer for enterprise decision-making, especially as the company expands across the US and Europe.
With a $28M Series A led by a16z and clients across the US and Europe, Hilbert is now trying to turn that category argument into enterprise infrastructure. Its team spans Istanbul, San Francisco, Barcelona, and London, and its client base includes Walmart, giving Hilbert a place inside the kind of large-scale retail environment where growth systems must handle complexity rather than avoid it.
Still, Akaoglu’s long-term focus is not only market expansion. He is interested in the abstraction problem: how to make AI work across hundreds of companies with different data structures, customer definitions, internal processes, and edge cases.
“Making AI work for one company is hard,” he says. “Making it work across many companies, where the same words can mean different things in different data systems, is the real challenge.”
That challenge sits at the center of Hilbert’s future. The company is expanding into additional problem domains for enterprise teams, moving beyond its current scope while keeping the same foundation: AI should help businesses decide, execute, and learn from what happens next.
Akaoglu sees that as the technical legacy worth building. Not AI that produces more analysis. Not AI that leaves teams with another tool to check. AI that becomes part of how a company learns to grow with more precision over time.
“The point is not to add more noise to the business,” he says. “The point is to build a system that helps the business compound on what it learns.”
For companies already overloaded with data, that may be the real shift. The winners will not be the ones with the most reports or the most disconnected experiments. They will be the ones that can turn learning into coordinated action.
That is the future Hilbert is trying to build.
“Growth should be provable,” Akaoglu says. “It should be something a company can understand, act on, and improve over time.”
For more information on Ozgur Akaoglu, visit the Hilbert website.
