Image credit: Supplied

With the news that MongoDB has announced plans to expand its footprint in the Middle East, following five years of growth in the region, it seemed like a perfect time for The Byteline to chat with (one of) the brains behind the innovative company.

MongoDB’s unified, intelligent data platform was built to power the next generation of applications and is the most widely available, globally distributed database on the market. The company expects to more than double its headcount in the region this year, with a focus on hiring in sales, solutions architects, channel partnerships, and customer success roles.

The investment comes after MongoDB’s strong performance, with the company expanding rapidly in the Middle East and North Africa at the same pace as its global growth since its regional launch in 2019.

With the UAE and Saudi Arabia aiming to become some of world’s most advanced nations in Artificial Intelligence (AI) and advanced technology, The Byteline sat down with Boris Bialek, Vice President and Field Chief Technology Officer, Industry Solutions, MongoDB, to discuss the company's recent developments and vision for the future.

‍

AI adoption is moving at an unprecedented pace, and many enterprises struggle with fragmented data infrastructure and outdated systems. From your perspective, what are the biggest bottlenecks in AI acceleration, and how can modern databases help businesses overcome these challenges?

It's a very simple picture. Historically, data was static. You loaded data into a database, kept it there, and ran end-of-day processing. But those systems don't reflect today's needs. AI is about real-time decision-making. It needs to understand, "What is my desire right now?" For example, if I'm at a store—whether physical or digital—it doesn’t help to know that I bought nappies last week, especially if I don’t have kids. AI needs a completely different approach to data.

You need real-time capabilities to mesh streaming data, like tracking mouse movements on a website. If someone hovers over a product they've never bought before, you could offer them a 20% discount in real time. Historically, systems were fragmented—separate order baskets, delivery systems, and supply chains. AI requires lifting data to a higher level to make it actionable.

This applies to financial services, too. Traditionally, banks had mortgage accounts, savings, investment plans, and checking accounts that weren’t integrated into a single view of the customer. AI changes that, making security a bigger challenge. Security in silos was easy because data was so scattered that no one knew where it was. But when you bring it all together, you need encryption and access controls. This is why we’ve been so successful in the UAE.

And when it comes to data sets and scalability, where does MongoDB fit in?

Our performance is unbelievable. Think of the document model as a folder containing a piece of paper with everything you need on it. In contrast, a relational database is like a box of index cards—you have to sift through multiple cards to get the data. That approach was smart for library systems, but it makes no sense today.

With MongoDB, instead of looking at "Boris" as just a name in a table, we can store Boris and his accounts in a single document, making data retrieval much faster. In traditional banking, one full transaction can involve 15-20 database interactions, with each transaction requiring access to multiple tables. That means millions of index lookups, slowing down processes significantly. We've seen real-world cases where MongoDB reduced query times from 80 milliseconds to just 1 millisecond. This makes a huge difference, especially for real-time fraud detection.

Today, fraud detection needs to happen in 50 milliseconds or less. If a fraudulent transaction isn’t flagged instantly, the money is gone. We worked with Networks International to move from in-process fraud detection to upfront detection. This means transactions are analyzed before they are approved, thanks to MongoDB’s real-time data processing.

That’s a game-changer. And how does MongoDB help companies adapt to evolving data regulations globally?

This is a bigger question than AI—it’s about data sovereignty. Governments are prioritizing national data protection. This is happening in Europe, in the UAE, and beyond. We’re pleased to work with Moro Hub, a data sovereignty provider in Dubai, to ensure that the UAE has its own infrastructure.

MongoDB also delivers industry-first encryption solutions. We encrypt not only at rest and in transit, but also during processing. That means even if someone accesses the application, they won’t see decrypted data. This level of security is critical for AI and global compliance.

That brings me to integrity. Some AI companies have high-profile figures leading them—Elon Musk, for example. How does an AI company remain unbiased?

Historically, everyone was excited about general-purpose LLMs like ChatGPT. But now, there’s a shift toward enterprise AI. General LLMs are not optimized for niche tasks, like insurance claim settlements. For AI to be truly useful, businesses need to train their own models with their own data.

For example, in insurance, claims data has been hidden in EDMS systems for years. By making that data actionable, companies can train AI models to process claims without external data. However, general-purpose LLMs are excellent at summarization—like condensing 50 research reports into a digestible format.

And how do smaller companies compete with big tech giants like AWS and Microsoft?

By offering better solutions. Microsoft has tried to compete with us by mimicking our database approach, but clients still choose MongoDB because it’s the original. Businesses don’t want vendor lock-in. If you only have one horse to ride, you have no options. But if you have three, you can pick the fastest one.

Every major cloud provider claims their LLM is the best, but no client wants to be locked into one vendor for life. The real value lies in corporate data, and businesses want to keep control of that.

AI is evolving rapidly. Where do you see it in 12 months?

If I knew for sure, I’d be a billionaire! But I see more standardized solutions emerging. Companies like Cognigy, Observe AI, and Encore Cloud AI are creating easily deployable AI solutions. There’s also a major shift toward AI-driven automation. We recently helped an insurance company implement an AI-based claims system. Now, customers submit pictures of their car damage and get approvals within minutes instead of weeks.

"If I knew for sure, I’d be a billionaire!"

In retail, AI can reduce return rates by recommending the correct shoe size in real time. In agriculture, predictive maintenance allows tractors to order replacement parts before they break down. The potential is massive.

MongoDB certainly has made big strides in a short time.

We started in Dubai in 2019 with a small team, and now we’re working with major banks and retailers. We even helped build the UAE’s COVID registration app in just two weeks. It’s amazing to see how far we’ve come.

‍