Engineering excellence: The power of continuous learning

“Is this your very best effort?” said Bob, as I handed him my report whilst gently pushing the document back towards me. It was 1992, a confidence challenge and a pivotal moment in my career. Bob was my line manager in my very first job as a new graduate. I pulled the document back and spent a few days improving the content, asking peers for review and sleeping on it. I was 22 and the paper was important. It was about electrical protection characteristics on the interface of some high-technology telecoms kit. It was a time of great technical innovation, which laid many of the foundations of the internet. A few days later I was rewarded with a “very good”.

The learning moment worked. It was about me having confidence in my own ability, evaluating my own performance, and seeking help when needed. It can take a while for engineers to consistently reach this point and requires the humbleness to recognise and acknowledge their own competency gaps. This is an essential part of being a competent engineer and I’m sure this will be recognised by others reading this article.

During my career, there have been a number of innovation waves which I have had the privilege of being associated with. High-capacity digital networks, mobile, broadband, Wi-Fi, voice over IP, the smartphone, the app, the cloud, SaaS, and the data centre becoming the foundation of this technological evolution.

I think we’re now in the midst of the next era of change, AI infrastructure and it’s seismic. There are common themes through all these technology shifts. Demand for new competencies, too few resources, struggling supply chains, infrastructure in short supply, customers with funding and motivation, but finding it challenging to keep the right focus on the key outcomes.

The AI era has triggered a significant shift towards a more strategic and risk aware approach for data centre operators in these key areas:

Sustainability

The extraordinary demands for power and the resulting potential environmental impact means that the focus on sustainability needs to be sharper than ever. A much deeper understanding of this issue is needed across all roles in the facility. This is particularly acute at the design stage as small gains in efficiency will result is large energy savings and lower environmental impact.

Power Density

The increases in power density and extensive use of liquid cooling will require increased technical competency. Aside from safe working, the potential impact of human error or equipment failure can be much greater than traditional design. Operations team members in the facility must perform comprehensive risk assessments before any work and have the right skills to do those audits. Procedural accuracy and mitigation of risk is more critical than ever before.

Criticality

Despite the number of racks within an AI platform, it is, at the end of the day, a single high-performance machine. Failure of just one rack can have a range of IT consequences from throttled processing to job failure and restart. All of which may have very meaningful commercial outcomes  for the user of a data center. They may have designed highly resilient IT architecture, but it is the data centre operator that has the responsibility to deliver ultra-reliable supporting infrastructure. This commercial awareness and the association with risk management is now a fundamental tenet of a strong data centre operations team member.

Today we have moved beyond simply maintaining infrastructure, to ensuring operations aspects are considered from the very beginning of design, how those complex interconnected systems are accepted by operations (not “handed over to operations”) and how they will be managed in-life. Throughout all of this, we need to understand the implications for asset maintenance and incident management. This requires an understanding of not just operational process, but also the underlying technology, system level behaviour, the context, and the tenant’s approach to the white space.

At Polar we’ve invested heavily in new site-based resources, prioritising the onboarding of new engineers and technicians from diverse disciplines. This includes valuing those that come from more mission-critical environments than data centres, and recognising their practical skills and mindsets. This isn’t just about filling vacancies; it’s about cultivating a team ethos of proactive and shared learning.

This positive story comes substantially from our site FM leadership team. I’ve witnessed first-hand how those with more skills in a particular technical discipline are teaching those with less experience. What’s really interesting is that this is happening organically; no pressure is being applied, and there’s been no demand for mentoring. There’s a motivation for the whole team to learn and become fully competent as quickly as possible. We’ve gone beyond simple training provision and cultivated a culture that actively encourages knowledge transfer, experiential learning and a willingness to embrace the challenges and complexities in an AI data centre. There are “Bob” style learning moments happening in our team regularly and the genuine levels of confidence are rising daily. I wonder how many other AI data centre operators can say this.

This experience is a powerful illustration of how both technical and leadership proficiency can unlock a deeper understanding of the potential challenges ahead and foster comprehensive risk mitigation. I believe this is a direct consequence of one of our key values “To have a problem-solving, and collaborative mindset that’s anchored on equality and transparency.” The outcome of this is better operational performance in this new frontier of AI-ready data centres, which is great news for Polar, our people, and our customers.