Why data center skills are becoming critical for AI growth

AI and data center growth is accelerating, with over €100 billion ($116 bn) in European investment reshaping the data center landscape. But with huge growth comes responsibility. European data center operators risk overlooking the critical skills required to keep data centers operational.

Infrastructure skills must not fall behind growth. Yet building and keeping these skills is a challenge, with almost two-thirds of operators reporting problems retaining employees, finding qualified staff, or in some cases both. AI growth is hugely exciting, but data center operators risk long-term failure if they cannot build and retain the skills required.

Here are five ways data center operators can build the infrastructure skills that AI needs.

Don’t dismiss cabling as “legacy”

As AI transforms data centers, vital infrastructure skills such as cabling and racking risk being overlooked as “legacy.” Yet without skilled engineers, the digital backbone AI depends on will falter.

Too often, cabling is treated as an afterthought, with engineers told to “just pull a cable” while designs prioritise hardware, power, and cooling. This short-term focus ignores future growth. With hardware cycles accelerating – new systems from Nvidia and others emerging every few months – cabling installed for current needs quickly becomes inadequate.

To avoid this gap, infrastructure skills must be recognized as skilled trades, requiring precision, adherence to standards, and collaboration. Much like carpentry, success depends on craftsmanship and coordination to ensure resilient, long-lasting digital infrastructure.

Putting foresight front of mind

When retrofitting sites or installing the latest GPUs, engineers must understand exactly what each cable does. How does it interconnect? How can changes cause disruption?

This level of understanding demands foresight in design, as very few operators leave space or ways for future updates. This creates bottlenecks when retrofits become unavoidable, requiring new hardware, cooling, or cabling layouts that the facility was never designed for. The skill and experience of cabling engineers here is vital. Their ability to pull, switch, and relabel without disrupting live environments ensures continuity.

Engineers must be trained to adapt layouts, validate systems, and manage live changes. Investing in infrastructure and people in this way creates long-term ROI and avoids costly cycles of retrofitting, while also developing a workforce that can support AI’s evolution.

Repositioning infrastructure skills as central to AI development

The success of AI is reliant on the infrastructure foundations on which it is built. But too often, infrastructure roles like cabling are seen as low-status roles, or a stopgap before “better” roles in the IT or data center industry.

Cabling, racking, and on-site deployments are undervalued. They’re typically last in line for budgets and recognition. This perception undermines recruitment at a time when operators face an acute talent shortage.

This perception must shift, and operators need to view and talk about infrastructure roles as being vital for AI development and progress. Without robust cabling or racking, data cannot move at the scale AI requires. Building pride is essential: infrastructure professionals are the enablers of the digital economy. Recognizing and celebrating this will help attract the next generation of talent.

Train, retrain, and train some more

Training and retraining with AI environments should become a key piece of the daily workflow. This will allow skills to evolve in parallel with infrastructure. Apprenticeship-style models, where junior engineers learn directly from experienced colleagues, remain one of the most effective ways to transfer practical knowledge and instill discipline.

The use of digital tools can aid this process. Digital twins, AR headsets, and SOP-linked dashboards provide guidance in real-time during live deployments. This reduces errors, improves safety, and shortens learning curves. The use of design platforms and AI agents can also demonstrate best practice by modelling cabling layouts and power distribution.

The right balance is essential: hands-on craft skills, combined with digital literacy and AI-augmented tools. Only by embedding training into day-to-day work can we build a workforce ready to manage tomorrow’s infrastructure challenges.

Safeguard the talent pipeline

There is a persistent talent shortage facing all data center operators that will require the industry to hire an additional 300,000 staff. An ageing workforce also means a chunk of our skilled labour could retire at the same time, leaving operators short on expertise. Compounding this is the fact that every operator is competing for the same limited pool of workers.

This means we must safeguard the pipeline. What’s required are apprenticeship schemes, vendor partnerships, and internal academies to help grow skills internally. So too is retraining from adjacent sectors, such as the armed forces. Operators should also look to early engagement with schools and universities to inspire the future workforce. Another avenue is tapping into neurodiverse talent and widening access to underrepresented groups to broaden the skills base.

Careers should be framed as valuable and long-term. These are infrastructure roles in fast-growing markets with job security tomorrow and for years to come – offering transferable skills, entrepreneurial routes, and the pride of being modern digital key workers.

Building the workforce for tomorrow’s digital infrastructure

Building skills is an industry-wide issue that demands collaboration. This means collectively recognizing cabling and racking as skilled trades essential to reliable AI performance, reframing infrastructure roles as vital to AI progress, embedding training into daily operations, and safeguarding the talent pipeline with apprenticeships, retraining, and clear pathways.

Trusted partners, vendors, and the wider ecosystem all have a role to play in developing and providing the skills, training, on-site delivery, and long-term services needed to ensure critical expertise evolves alongside the technology it supports. Only by investing in people, process, and partnerships can we build the workforce needed to power the next era of AI-ready infrastructure.