How to build a world-class AI team? 

Wojciech Zaremba, Jakub Pachocki, Szymon Sidor, Filip Wolski, Aleksander Mądry. These names might not be household conversation starters, but in the world of artificial intelligence, they represent some of the brightest minds behind groundbreaking technologies. The Chief Scientist at OpenAI, a co-founder driving research behind ChatGPT, the head of AI safety – all sharing one unexpected common thread: their Polish origin. 

Their prominence reflects a broader pattern. Polish experts consistently secure leadership positions at innovative companies pushing technological boundaries At OpenAI alone, Pachocki leads as Chief Scientist, while Zaremba, one of the co-founders, drives research underpinning technologies like Copilot and ChatGPT. Mądry heads the critical Preparedness team, focusing on AI safety. Their success suggests that when building world-class AI teams in 2025, business leaders would do well to look beyond traditional tech hubs. 

“I don’t know what Poland does as a culture and system to create such incredible engineering and research talent…but it’s quite remarkable the impact that it’s had.” – Sam Altman, OpenAI CEO, during a visit to Warsaw University, 2023 

This concentration of Polish talent in pioneering AI roles raises an important question: what makes Poland such a fertile ground for AI expertise, and how can organizations tap into this potential? As businesses race to build their AI capabilities in 2025, understanding this phenomenon could provide a crucial competitive advantage. 

The evolving demands of AI development 

While stories of AI pioneers capture imagination, the reality of AI development in 2025 demands far more than cutting-edge technical expertise. Success requires mastering an interconnected web of competencies spanning machine learning, ethics, and operational excellence. 

But here’s what’s interesting: technical expertise, while essential, isn’t the main challenge companies face when implementing AI. Industry analysis reveals a more complex picture – successful AI engineers today need to master a complex set of competencies spanning machine learning, data engineering, MLOps, and even ethical AI governance. 

This becomes clear when we examine the state of industry in 2024. Organizations using AI in at least one business function have jumped from 20% in 2017 to 72% in early 2024 (McKinsey State of AI Report, 2024).  

To do so, naturally, businesses needed a skilled workforce that not only had the experience but was also trained in relatively new competencies. And the road from employing the right people to having working AI solutions is a rocky one. Think of the full lifecycle of an AI project: data pipelines, security measures, integration and scaling. It’s why technologies like Docker and Kubernetes have moved from nice-to-have to essential – they’re the building blocks of reliable, enterprise-grade AI solutions (DataCamp, 2024). 

AI success, understood in those complex terms, also relies on understanding the specific challenges of each industry. A brilliant AI solution for healthcare might fail completely in finance or manufacturing, not because of technical flaws, but because it doesn’t account for the unique demands and regulations of each sector. The most effective teams blend their technical expertise with deep industry knowledge, so it’s not only a matter of technical skills but also domain-specific expertise. 

At the beginning of 2025, no business is unaware of how pressing investing in AI solutions is, but there is still room to grow when it comes to building actual frameworks and best practices. However, as organizations rush to build their AI capabilities, they’re encountering a new set of challenges. 

Current obstacles and constraints 

Despite this universal push toward AI adoption, a sobering reality emerged in 2024: knowing you need AI talent and successfully building AI teams are two very different challenges. Last year’s numbers tell the story – while AI hiring demands jumped 67%, the qualified talent pool grew at just a quarter of that rate (McKinsey State of AI Report, 2024). The market’s response was swift but insufficient. The global AI education sector hit nearly $6 billion (Grand View Research 2024), with non-technical professionals rushing to build AI skills through platforms like LinkedIn Learning, where enrollment spiked 160% (Microsoft and LinkedIn, 2024). As we move deeper into 2025, most businesses find themselves caught between mounting pressure to deliver AI solutions and a persistent talent shortage. 

The impact varies dramatically by location. Major tech hubs have turned into expensive battlegrounds for AI talent, with rising salary demands straining even established companies. Emerging tech markets face an even tougher reality – their best engineers often leave for bigger markets, creating gaps that ripple through entire regional economies. Beyond technical skills, companies need people who understand both AI systems and specific business domains – a combination that’s proving remarkably rare. 

This scarcity manifests differently across various markets. Established tech hubs see fierce competition, with businesses locked in costly bidding wars for experienced engineers. Meanwhile, emerging tech centers struggle to retain expertise, creating a brain drain that affects entire regional ecosystems. The issue extends beyond finding people with technical prowess – companies need professionals who understand both advanced computing and specific business domains. 

Compensation, predictably, sits at the heart of the matter. Senior AI specialists’ salaries have risen by over 35% in the past two years (PWC AI Jobs Barometer, 2024), forcing many businesses to rethink their recruitment approach. Investing in AI experts was simply too costly, not to mention that the high pay alone doesn’t guarantee success. Many well-funded projects still falter when groups lack the right blend of experience and sector knowledge. 

But securing skilled teams is just the first challenge. The pace of AI advancement has created a constant pressure to adapt. Solutions that worked in early 2024 were often completely outdated by year’s end. Today’s AI professionals face an ever-growing list of responsibilities. They need to master new LLMs, learn emerging frameworks, and revise deployment strategies. All this while maintaining their existing projects. Many organizations now struggle with a fundamental question that goes beyond their budget concerns. How can they help their teams stay current without compromising ongoing work? 

As we move through 2025, successful AI implementation looks less like a race and more like a strategic puzzle. Companies are learning that brute force approaches – whether through massive hiring budgets or rapid tech adoption – often stumble. What’s emerging instead is a new focus on building AI capabilities systematically, where both teams and projects can grow without burning out. 

Building solutions that work 

At ITSG, we’ve seen the AI talent challenge from both sides – as a Polish tech hub working with global clients, and as practitioners building AI solutions ourselves. While there’s no perfect formula that works for everyone, our experience has shown what makes the difference between teams that struggle and those that succeed. 

The most successful AI teams we’ve worked with share three simple but powerful approaches: 

  1. Look beyond technical skills. Yes, coding expertise matters – but the best engineers are those who can understand what a business really needs and find practical ways to use AI to solve those problems. When we evaluate talent, we specifically look for people who can explain complex ideas simply and turn technical possibilities into real solutions. 
  1. Share knowledge constantly. Instead of letting each team work in isolation, we encourage regular sharing of ideas and challenges. Simple things like team workshops and pairing experienced developers with newer ones make a big difference. Teams learn faster, solve problems quicker, and build better solutions when they regularly exchange insights. 
  1. Understand the industry. An AI solution that works perfectly for one industry might completely miss the mark in another. That’s why we make sure our teams deeply understand not just the technology, but also how our clients’ businesses work. This combination of technical and industry knowledge consistently leads to better results. 

These principles are great in theory, but how do we actually put them into practice? We’ve developed a straightforward approach that starts with understanding exactly what you need and ends with a fully operational team – typically within 3-8 weeks. 

And we start by listening. No generic solutions or one-size-fits-all approaches. We take the time to understand your specific challenges, whether it’s integrating AI into existing systems, ensuring security and compliance, or building new AI capabilities from scratch. This careful groundwork helps us build teams that can address the most common challenges companies face: 

Challenge ITSG Global Solution Benefit 
High IT Costs Transparent pricing with up to 35% reduction Significant budget optimization 
Talent Shortage Access to Poland’s elite IT talent pool Immediate skill gap closure 
Security/Compliance issues Expertise in European data protection laws and security-first approach Secure and compliant IT solutions 
Legacy Systems Specialized modernization expertise Enhanced competitiveness 
AI and Innovation Gaps Cutting-edge Generative AI capabilities Accelerated innovation 

Whatever stack you need, we can build a team to match. Need IoT platform developers or a team to modernize legacy systems? We’ve done that. Looking to add .NET, Java, React, or Node.js expertise? We know where to find the right people. Recently started exploring AI? We’ll put together experts who understand both the technical requirements and your business context. But what matters most isn’t the list of tools and frameworks. It’s creating teams that understand what you’re trying to achieve and know how to get there.

This brings us back to where we started. Remember those Polish engineers making waves in global AI? They represent something bigger than individual success stories. They’re part of a tradition where technical excellence meets practical problem-solving, where innovation serves a purpose rather than just making headlines. 

References

  1. DataCamp (2024). “The 14 Essential AI Engineer Skills You Need to Know in 2025”
  2. Grand View Research (2024). “Artificial Intelligence (AI) Education Market Report”
  3. McKinsey & Company (2024). “State of AI Report 2024”
  4. Microsoft & LinkedIn (2024). “Work Trend Index: The State of AI at Work”
  5. PwC (2024).”AI Jobs Barometer: Global Trends in AI Employment”
  6. The Recursive (2023). “OpenAI CEO Sam Altman: ‘I Don’t Know What Poland Does to Create Such Engineering and Research Talent'”
  7. Velocity Media (2024). “Essential Skills & Trends for AI Engineers in 2025” Wired (2024). “8 Google Employees Invented Modern AI: Here’s the Inside Story”

Join our newsletter

Subscribe to the newsletter to stay updated with the latest industry news
and our activities such as blogs and events!