How Progress in Power Equipment is Moving Civilization Forward – An Interview With Leonel Clemente

Written by: Henry Van Niekerk

---

According to estimates from the International Energy Agency, investments in new data centers will surpass $580 billion in 2025, for the first time exceeding global spending on oil exploration and processing infrastructure. The primary force behind this shift is the rapid development of artificial intelligence. To support AI’s growing computational demands, new data centers are being built worldwide, becoming the backbone of the emerging digital economy. Yet at the core of this massive infrastructure lies something entirely physical, switchgear, circuit boards, and the electrical systems that keep data centers alive. Leonel Clemente, head of the PDUs & Switchgear division at one of the world’s leading technology companies, has been at the forefront of building the modern digital landscape. Working within cutting-edge, high-tech environments, he developed new-generation equipment under sterile conditions, hardware that laid the foundation for today’s data-center architecture. In this exclusive interview, Leonel Clemente speaks with Brainz Magazine about his innovative approach and how it helped accelerate the arrival of the digital era.

Leonel Clemente, Operational Leader in Critical Power Systems & AI-Driven Manufacturing

Leonel, you developed a unique approach to designing and manufacturing power-distribution equipment for AI-ready data centers. What makes it so remarkably effective? 

My approach was shaped during my time leading the corporate manufacturing sites in McKinney and Plano, Texas. I was responsible for building a production model for Low and Medium-Voltage Switchgear, as well as overseeing the manufacturing of high-capacity PDUs, including 500 kVA and 1,060 kVA units, ATS (Automatic Transfer Switches), and RPPs (Remote Power Panels) for mission-critical environments.

When I joined Maverick Power, the initial plan for the new facility I would be heading was to produce around 60 Switchgear sections per week. For a 200,000-square-foot plant, that would have been a respectable output. However, once I began working on the project, it became clear that we could significantly expand this capacity by rethinking equipment scheduling and restructuring workflow organization.

In many industrial sectors, efficiency-boosting algorithms have already been developed and proven effective. But in the Switchgear, PDU, ATS, and RPP segments, such frameworks simply do not exist. These systems have become increasingly complex, and explosive demand has emerged only in recent years. With no ready-made solutions to draw from, I designed my own method for gradually increasing production volumes using the existing manufacturing infrastructure.

Switchgear production has its own specific nature, you cannot simply ramp up output overnight because a manager wishes it. The same applies to PDUs, ATS, and RPPs, which require precise engineering, safety protocols, electrical validation, and strict regulatory compliance. I developed a detailed schedule and began scaling output step by step, strictly following all manufacturing standards and certification requirements. As a result, the facility doubled its production, reaching 120 sections consistenly per week, while successfully delivering high-complexity Switchgear, PDUs, ATS, and RPPs for some of the most demanding applications in the industry.

In the critical-power sector, nothing is more important than On-Time Delivery (OTD). Unlike traditional manufacturing environments, where minor delays can often be absorbed, data-center power infrastructure operates on rigid, non-negotiable timelines. Every switchgear assembly, RPP, ATS unit, and power-distribution component is tied to synchronized installation schedules involving electrical contractors, commissioning teams, utility providers, and hyperscale operators.

If power-distribution equipment does not arrive precisely on time, the entire data-center project is forced into delay, servers cannot be installed, HVAC and cooling systems cannot be activated, commissioning windows collapse, and revenue-generation dates shift, often resulting in multimillion-dollar impacts for operators.

For global clients such as Equinix, NVIDIA, Google, AWS, and Microsoft, delivery precision is as critical as engineering quality. These companies depend on flawless execution. A manufacturer’s real competitive advantage lies not only in designing robust equipment but in meeting every milestone with absolute punctuality.

This is why OTD is widely recognized as the defining KPI of the critical-power industry. Under my leadership, lean workflows, predictive takt-time scheduling, equipment utilization modeling, and digital traceability systems were integrated into daily production rhythms. These systems enabled us to achieve best-in-class reliability, maintaining a 99% OTD performance during the high-volume deployment of low- and medium-voltage switchgear and related power-distribution systems for AI-driven data-center programs.

Engineering delivers capability; On-Time Delivery delivers trust. And in the critical-power business, trust is the true currency that determines which organizations are chosen for the world’s most demanding infrastructure projects.

During this period at Maverick Power, our largest customers included Crusoe, whose AI-driven data centers require extremely high-density power architectures, and Yondr, one of the world’s fastest-growing hyperscale developers. Serving such clients demanded flawless execution not only in engineering and production, but in materials readiness. In critical-power manufacturing, it is common to have millions of dollars in transformers, breakers, bus bars, and switchgear assemblies ready for production, yet a single missing washer, bolt, or bracket, often valued in cents, can halt the entire assembly line.

This dependency on perfect material alignment reveals a fundamental truth about the industry, high-cost components do not keep production moving, complete components do. A simple oversight in a Bill of Materials (BOM), an undelivered fastener, or a mislabeled hardware kit can freeze output despite the availability of every major part. For hyperscale projects, where switchgear delivery defines the energization schedule, this level of precision becomes non-negotiable. Ensuring that every component, down to the smallest washer, is present, verified, and staged at the exact moment of assembly is as critical as the engineering work itself.

What did you rely on while executing such an unprecedented project? 

My experience played a central role. I spent more than 10 years at Seagate Technology, the global leader in data-storage systems, which in 1998 held nearly 35% of the global market. As head of the engineering department, I oversaw the technological design and development of Factory of the Future (FoF) production lines in Seagate’s Class-100 cleanrooms.

Back then, I found myself pioneering many of these processes. No other company had technologies comparable to Seagate’s, and there were no established methodologies for optimizing such high-precision environments. Still, I developed an innovative approach to managing manufacturing operations in cleanrooms. Ultimately, my solutions reduced energy consumption by 10% and increased productivity, on some lines by as much as 20%. It’s important to understand that all of this work was carried out in conditions that were essentially sterile. That experience taught me that nothing is impossible, even the most ambitious task can be solved if you approach it comprehensively and without haste.

So when I saw the untapped efficiency potential at the Maverick Power facility, I took ownership of the project and delivered it. It wasn’t easy, but it was absolutely achievable.

Today you lead the switchgear division at another major technology company, where you’ve become the driving force behind a different kind of innovation, scaling the low-voltage (LV) and medium-voltage (MV) switchgear business. How can this impact the development of data-center infrastructure?

In the most direct way possible. The division I lead supplies critical equipment to giants such as Equinix, NVIDIA, Google, Amazon Web Services (AWS), and Microsoft. I oversaw the launch of a new switchgear manufacturing facility at our Dallas plant. The site opened in early November, and the first deliveries have already been successfully completed. In previous roles, I focused on improving efficiency at existing manufacturing operations. This time, however, I built the project from the ground up, integrating lean manufacturing principles, predictive maintenance, and full digital traceability from day one. In effect, we created an entirely new production framework for manufacturing high-tech electrical equipment. This approach will allow us to scale output more efficiently and help meet the rapidly growing market demand for switchgear, a key element in building the infrastructure required for modern data centers.

Your professional path reflects industrial evolution, proving that the success of technology depends on the strength of the systems that support it. How do you anticipate the future and understand what conditions will be required for the data centers of tomorrow? 

The modern success of AI technologies is driven by a global shift in something entirely physical, the true technological transformation is etched in metal, wire, and voltage. Power generation is not an auxiliary act in the artificial-intelligence revolution, it is its driving force. This trajectory became clear a couple of decades ago, when the first data centers emerged. It was then that industry professionals realized, organizations capable of developing smarter, faster, and more resilient power-distribution systems would not simply keep pace with artificial intelligence, they would define its direction. In the age of AI, the future belongs to those who know how to build it from the ground up.

In an era of hyper-computing, the most important task is to prevent overheating of the very physical systems that enable all this work. That is why energy-efficient circuit breakers now play a critical role in data centers. The breakers themselves are evolving too, becoming more technologically advanced and beginning to use AI support in their own operation.

As data-center capacity grows, I foresee a rising demand for greater energy efficiency, because such facilities require enormous amounts of power. Development is already underway on systems capable of operating with renewable-energy sources. Looking ahead, even switchgear housings are beginning to be made from recyclable polymers. All these challenges, in turn, require PDU manufacturers to rethink their production technologies in order to integrate the solutions the market now demands. At the same time, we must maintain production volumes and impeccable quality, because the performance and safety of data-center operations, and the integrity of the data itself, depend directly on the reliability of our systems.

You consistently achieve truly breakthrough results by combining advanced engineering, operational excellence, and strategic business leadership. What’s your secret? 

I’m a realist, and I understand well that in our industry progress moves so fast that a solution developed today can’t simply be reused a few years later at another facility. So I chose a different path, creating principles, frameworks, and scalable methodologies that remain relevant for decades. My core mission is to design solutions that ensure reliability and resilience for critical power infrastructure. Today, this infrastructure is what enables technological progress for all of humanity, and that responsibility shapes everything I do.

Follow me on LinkedIn for more info!

Read more from Leonel Clemente