Powering the AI Era: How PPC Is Engineering the Grid of Tomorrow

Far from a passing trend, artificial intelligence (AI) has rapidly proved itself as a load-bearing pylon in power engineering. Like the rest of the continent, Southeastern Europe is currently racing toward clean energy targets and massive digital transformation. At the center of this transformation is the grid itself, where managing operations now requires processing vast amounts of telemetry data in real time, from smart meters in Romania to regional hyperscale infrastructure.

We sat down with Daniel Albu, who leads PPC’s data and infrastructure team. At PPC, digital solutions and data architectures are embedded directly into physical operations, driving everything from predictive maintenance on high-voltage lines to large-scale renewable integration.

Below, Daniel breaks down how PPC handles heavy telemetry, why developers need a “resilience-by-design” mindset, and what’s next at the intersection of clean energy and AI.

From your perspective leading PPC’s data and infrastructure, what was the tipping point where AI went from an experimental pilot to a heavy, high-stakes engineering workload in production?

AI started as a tool like any other – a powerful tool, but a tool nevertheless. The high stakes emerged the moment we could actually measure the benefits of its use, which didn’t take too long.

Today, AI transformation is a top group priority at PPC, touching every system and process from top to bottom, across the entire energy value chain – energy generation, trading, distribution and retail. 

Take our operations in Romania, for example. In distribution, predictive maintenance algorithms powered by historical network data are more than 97% accurate, signaling areas at risk of damage before failures occur, while drones flying over high-voltage lines use image recognition to detect defects before they affect customers. Over two million Romanian customers now have smart meters that transmit consumption every 15 minutes, data that powers tools such as the My Energy Coach virtual assistant in our myPPC app.

At the group level, our flagship example of AI and digital infrastructure managing an intermittent renewable energy portfolio at scale – in live operation, not in a lab – is the €5.75 billion project to transform the former West Macedonia Iignite mining area in Greece into a clean energy and digital infrastructure hub. The project combines a 300 MW hyperscale data center (scalable up to 1 GW), alongside 2.13 GW of solar capacity and 1.1 GW of storage.

Simultaneously, in Romania, we have rolled out enterprise AI tools (like Claude Enterprise) to an initial wave of approximately 1,000 employees as part of our responsible AI adoption program.

Our theme for How to Web Conference 2026 is “What will you build when you can build anything.” In a sector bound by physical grid limits, what does that freedom look like? If your team had no limitations for a month, what would you build?

I’m an engineer at heart – building things is in my DNA. However, even if AI brings the raw cost of development close to zero, it won’t radically change traditional corporate realities overnight. Instead, I think there will be an increase in delivery speed, but AI won’t do the strategic innovating for any of us, at least not yet. If I had no technical limitations, I would build whatever yields the most value in the shortest time to fund that expertise.

You see, there is this idealistic, but frankly unsustainable, notion in the market right now that you can actually build products with AI without deep technical domain knowledge. This is certainly dangerous because without the technical depth, there is no way you will be able to measure resources and understand the infrastructure costs. From a security perspective, I can’t even overstate how high those risks are.

How do you keep your data teams at PPC moving fast while staying deeply integrated with the physical grid?

We build directly alongside our business teams – the people managing physical assets corresponding to distribution and production segments, such as grids, photovoltaic panels or wind turbines.

The era of IT as a passive support function or isolated devshop is over. In PPC’s case, IT is part of the business and needs to be integrated across all departments. In fact, digital solutions will effectively be the business within the next ten to 15 years – no company will function without them, regardless of how advanced AI becomes. IT is digging its roots across the whole market, which is essential if we want to scale faster and stronger.

With the scaling of prosumers, the grid has evolved from a traditional one-way highway into a highly complex, two-way network. From a data-collection perspective, how do you architect pipelines to handle the massive flow of real-time telemetry?

We never had a technology problem with collecting or processing heavy loads of data. The real question is: what do you do with that data?

We have around two million smart meters, each generating around 17 distinct data files each daily. The peak load is registered around 4AM, which consists of approximately 800,000 meters. Our entire process of collection, processing and storage handles that entire 800,000-meter spike in about four minutes flat.

PPC is scaling up clean-tech and regional hyperscale data centers. How does your data playbook change when you are building the physical power foundation for Europe’s digital future?

Today, data is part of any capital investment and rooted in every decision we make. Every asset needs to be accounted for and monitored in real time, and for that, you need data.

To enable this, we are currently working on a set of mandatory requirements that will cascade down into all projects from the design phase, enabling us to collect and integrate everything from Day 1.

AI is both the energy sector’s hungriest consumer and its smartest tool for grid stability. Do you see this dual pressure as a conflict or as a natural loop that balances out?

It will inevitably balance itself out. The cost of energy and grid management versus the cost and benefits of AI will, at one point in time, balance out. Beyond AI, the energy sector is also racing toward reaching EU’s 2050 target of becoming carbon-neutral/energy independent and electrifying everything, which puts even more pressure on infrastructure. For sure the industry is changing fast, and companies will either adapt or disappear.

At PPC, this is what we’re focusing on – we’re designing systems for the operational realities five, ten and 15 years down the road.

When it comes to engineering modern energy solutions, what do you see as Southeastern Europe’s distinct tech advantages or talents?

Southeastern Europe hosts one of the strongest digital infrastructure and technical talent pools on the continent. What we need is a change of mentality: we are in the driver’s seat and need to work together in order to achieve an energy-independent Europe. We also need to step up our investments into regional deep-tech startups. We have historically lagged behind there, but things are finally shifting in the right direction.

The runaway AI boom is putting pressure on clean-energy commitments worldwide. Can we scale AI without compromising the clean energy transition?

They will scale together. Investments and developments are accelerating, but it’s not something that can be done overnight. We need to set a priority on what brings the most value per dollar and start from there. For instance, it makes little sense to scale AI while the core regional manufacturing sectors remain dependent on fossil fuels.

Let’s bring this back to the builders heading to How to Web Conference this October. When software developers leave standard consumer web apps to work in critical energy-tech, what is the number one mindset you wish they brought with them?

I have a saying: if you were dependent on your product, how would you feel about it? How would you feel if you lost power for two to three hours – no internet, no mobile networks? We would revert back decades in a matter of hours. 

So, whatever you build, you need to have resilience by design. You need to think about all possible scenarios and keep solutions up and running in the harshest operational environments.

Finally, let’s wrap up with a direct challenge to the crowd. If you could give just one piece of practical advice to the engineers and product builders walking into the venue, what would it be?

If you limit your identity to “I’m just a developer,” your role will be obsolete in the next five to seven years. AI might’ve made our jobs easier from one point of view, but a lot more complex from other standpoints. My advice? Try to adopt a business builder’s mindset, think outside the box and never stop learning.

Catch the future of energy tech live in Bucharest. Explore the lineup and secure your How to Web Conference ticket now.