An IT Entrepreneur from Tashkent Launched a Startup in the USA and Has Already Raised $2.7 Million

Akmal Paiziev from Tashkent has been in business for over 20 years. During this time, he founded five companies. He successfully exited two of them, selling the startups to major market players. Now, Akmal is developing a new project called Numeo AI, which creates specialized AI agents for logistics and freight shipping. Despite his extensive experience, this startup represents uncharted territory for the entrepreneur from Uzbekistan. First, the company is registered in the USA and focuses exclusively on the US market, while previously Akmal operated only in his home country. Second, the startup targets a different sector — it works directly with freight forwarders rather than other industry players. Third, the solution is based on artificial intelligence, a rapidly growing field. The project has already attracted several million dollars in investment from the venture capital firm NFX.

As part of the joint project 100 Startup Stories of Central Asia by Digital Business and Astana Hub, Akmal spoke about the development of Numeo AI and the specifics of the logistics industry in the United States, where the freight shipping market is already valued at $850 billion. We also touched on the industry’s skepticism toward artificial intelligence, and his plans to grow the startup across the U.S. market.

«There are many freight forwarding companies in the USA that were founded by people from Uzbekistan and other Central Asian countries»

— So, what were you up to before this? How did you end up in the IT world?

— I’ve been building startups for 20 years now. I’ve launched four major projects so far and successfully exited two of them. Numeo AI is our fifth company.

Earlier, we built a telemetry platform under the MaxTruck startup, the food delivery service Express24 (which was later acquired by Yandex), and MyTaxi, an aggregator that became a leader in corporate transportation in Uzbekistan. We also launched Workly, a platform for automating HR processes in companies. By the way, not long ago we announced the consolidation of assets with the Kazakhstani startup Clockster, which has now taken over the Workly product and team to develop it under a single brand.

Founders and business partners of Numeo AI: Beka Kodirov (CTO), Akmal Paiziev (CEO), and Bek Yusupov (COO) at the CABXPO'25 in Chicago

— You’ve worked in food delivery, passenger transportation, and telemetry — all of which intersect with logistics. In a way, Numeo AI feels like a natural evolution of what you were already doing. But still, what led you to this new project?

—I moved to the USA last year and started exploring ideas for a new company. The first one was totally different from what we’re doing now. We even began building it at one point.

At first, we wanted to focus on optimizing service in the restaurant industry. That startup was called SpeechSense. The idea was to create a kind of digital supervisor. In my previous company, I worked with restaurants for about seven years, so I knew their challenges pretty well. For example, management efficiency was often lacking, and staff communication with customers wasn’t always great. It quickly became clear that AI could really help in this area.

We wanted to create so-called digital microphones that could listen to and analyze how staff — like waiters and hostesses — interacted with guests. The AI would detect the context, pick up on complaints, recognize the tone of a customer’s voice, identify any issues that came up, and assess how the staff responded. Then, the system would offer recommendations on how to improve overall operations and, more specifically, customer service.

But we ran into a problem. If you want to build a big company, you have to sell your solution to major chains. And there was a major blocker: well-known franchises are very cautious about scandal and bad PR. Even just the idea that they might be “listening” to their customers could cause serious backlash. Theoretically, an article titled “Starbucks Is Listening to Your Conversations” could become a huge issue for the brand, even if the AI agent is fully compliant with US laws.

So this turned out to be a serious blocker for us. Legally speaking, all the issues in a case like this can be resolved. But if companies aren’t ready to adopt the solution because of branding or reputational concerns, it’s very hard to change their minds. We became fully convinced this was a real problem when we reached the stage of negotiating with some major restaurant chains.

So we put that idea aside and started thinking about where else we could apply our existing AI technologies and practices. Logistics came up as one of the potential areas. Everything is different here — there’s no eavesdropping involved, because the AI agent interacts directly with market players like freight forwarders, brokers, and logistics specialists.

Our idea is to optimize work that’s still done inefficiently, whether it’s slow or handled manually. For example, finding valuable cargo among thousands of offers and negotiating better shipping rates. That’s where artificial intelligence can already make a difference. It allows us to build fully capable digital employees.

What’s also important is that in the USA, there are many freight forwarding companies founded by people from Uzbekistan and other Central Asian countries. That makes it easier to access the market and find common ground with clients.

— You mentioned there are still a lot of inefficient processes in logistics. How does that typically work, and how does the AI agent help improve the situation?

— It all comes down to people — none of us is perfect. For example, we sometimes make emotional decisions that aren’t optimal. Or we act based on limited data, without seeing the full picture, and our assumptions or confidence are partly built on guesses.

Let’s imagine a situation. A truck driver is available and looking to take an order in Chicago. If you open the load board, you’ll see thousands of different cargo offers. How can anyone realistically analyze that much data in 20 minutes and figure out which option is the most valuable? And every second, new offers are coming in. That’s where algorithms do a much better job.

We collect big data, and our AI works with it. It understands the context, can negotiate, and advocates for the most beneficial terms for both logisticians and drivers.

«Many people are still afraid to start a conversation or even correspond with AI»

— What does Numeo AI consist of? What products do you offer, and what do they do?

— Our company creates AI agents for the logistics industry. In other words, digital employees. They help businesses operate more efficiently and make better decisions. Right now, we have two such agents:

1. Spot Finder helps shipping companies find better orders. For example, the AI can negotiate by phone or through messaging to secure better delivery rates for goods.
2. Updater Agent helps manage the entire delivery cycle after Spot Finder has identified an order and secured favorable terms for the carrier. The agent communicates with the driver, broker, and dispatcher, coordinating the process from start to finish.

— The shipping industry is massive, and I’m sure major players have already explored solutions similar to Numeo AI. Do you have any competitors?

— The segment we focus on — working specifically with freight forwarding companies — is still relatively open. At least for now, we don’t see any serious competitors. First, the AI space is still quite new. It started gaining traction about a year ago. Second, the niche we’re targeting is very fragmented and, as a result, quite complex. Many companies focus on the broker side rather than the shipping companies. And those that do target our segment usually lack the funding and resources needed to build strong solutions.

I don’t think Numeo AI will face serious competition in the near future. As of today, there are very few competitors.

— How many clients do you have today, and how did you land your first one? The product is new, there’s no direct competition — was it easy to convince companies to start working with you?

— Spot Finder is used by around a thousand companies, while Updater Agent is used by about 60 clients. As for negotiations, we didn’t face any major difficulties. For example, we secured funding even before the product was built — eight clients paid between $1,000 and $3,000. We had a clear understanding of their pain points and knew how to solve them. In turn, the companies saw that we understood the market and its specifics.

But we did face challenges when integrating the agents into existing infrastructure. There were both technical hurdles and resistance from employees when it came to adopting a new approach. The business owner might like the solution, but the team often resists anything unfamiliar — and gaining acceptance is never a simple process.

— What do you mean by resistance? Are people afraid that the AI agent might replace them?

— There are several factors. First, our product is new, and there hasn’t been anything like it in the logistics industry before. Second, there’s still a lack of trust in AI. Dispatchers can be cautious about using such an assistant. Cargo shipping is a fairly conservative industry, and many people are still hesitant to communicate — or even correspond — with AI.

«We weren’t actively looking for financing — the fund approached us first»

— What is the business model of Numeo AI?

— We currently use a subscription model, but we’re gradually transitioning to a different system and plan to take a percentage from each deal.

— Is Numeo AI making a profit yet? Any numbers you’re able to share?

— No, we are not profitable at the moment. We are still burning money. With our current client base of around 1,000 freight forwarding companies, we have not yet reached break-even. We have a team of 20 people. I am not able to disclose our MRR or any other financial figures at this stage.

— Speaking of financing, you raised funds from the venture capital firm NFX. Can you share some details?

— Yes, we raised a total of $2.7 million. We weren’t actively looking for funding. The fund approached us first. We quickly agreed on the terms, and the result was a funding round completed this May. However, for the US market, this is not considered a large amount.

— Do you have plans for another funding round?

—Not at the moment, but I think we may consider that opportunity in the future.

— Right now, you're focused on the US market. Do you have plans to expand into other countries or regions?

— No, we are not considering other markets at the moment. The U.S. market is huge, and there’s still a lot of work to be done here. The logistics industry alone is estimated at $850 billion. We help companies find orders on the spot market, which accounts for about 20% of that — roughly $170 billion. The BPO segment (Business Process Outsourcing, which involves transferring certain business operations to third-party providers - note by Digital Business) is valued at around $25 billion. So the market is more than large enough for us to focus on. Globally, there’s nothing quite comparable in terms of scale or opportunity.

— What development plans do you have for Numeo AI in the near future? For example, in 2026?

— It's very hard to make long-term forecasts. We’re a startup, and things can change quickly. Strategic planning is more typical for mature companies, while in early-stage projects like ours, we often act on an ad hoc basis.

— AI has exploded across so many industries lately. Do you think the hype will eventually settle down or is this just the new normal?

— The hype is definitely real, but I don’t think it’s going to fade. This momentum will continue. There are so many areas where AI can be applied, and we’re going to see a wave of new startups across all kinds of niches and industries.