By Matthew Hardman
AI gets the headlines. But the people who keep it running—system administrators — rarely get the credit.
Across Asia, as companies move from pilot projects to full-scale AI deployments, they’re hitting familiar roadblocks: messy data, incomplete records, and systems that can’t keep up. According to Hitachi Vantara’s latest State of Data Infrastructure Survey, 42% of organisations in Asia now say AI is critical to operations. Yet AI models are producing accurate results just 32% of the time. Only 30% of their data is structured.
Infrastructure isn’t failing. But it’s under serious pressure. Bridging the gap between AI ambition and execution often comes down to one thing: a well-managed data environment. That’s where system administrators come in.
These are the people managing exploding data volumes, securing systems, and keeping businesses online. They don’t make headlines, but they make AI possible.
As companies rush to automate, it’s easy to forget the human effort behind digital transformation. Sysadmins keep AI grounded in the real world. They deal with imperfect data, prevent outages, and ensure security. The seamless AI experiences we see rely on solid infrastructure and careful planning behind the scenes.
AI’s success depends on data quality and the reliability of the systems supporting it. That’s why sysadmins are critical. They keep pipelines flowing, fine-tune tools, and monitor for issues that could throw models off course. In fast-moving environments, this kind of precision is what keeps performance on track.
Without sysadmins, AI systems stall. Data pipelines break. Security gaps widen. As more businesses rely on AI for decision-making and automation, it is sysadmins who ensure it all keeps running. They are not just maintaining systems; they are building resilience into everything.
The real backbone of tech
Sysadmins are the architects of IT infrastructure. They manage servers, secure networks, deploy software, and solve problems most people never see. Their work keeps things moving and ensures teams can stay productive.
Think of them as the control tower of enterprise tech. They may not be flying the plane, but nothing gets off the ground without them. In the background, they are watching for turbulence, from server spikes to failed patches or breaches.
According to our survey, 71% of Asian enterprises are hiring talent with AI-related skills. Sysadmins are part of that shift. They ensure uptime, balance workloads, and optimise performance across distributed environments.
But their job is evolving. AI systems now run on graphics processing units (GPUs), tensor processing units (TPUs), and complex storage architectures. Sysadmins need to manage these resources and work closely with data scientists, DevOps, and compliance teams. Infrastructure is no longer just a backend function; it is central to strategy and execution.
And in today’s cloud-first setups, things scale fast. One misstep, such as a missed update or flawed configuration, can halt an entire AI pipeline. That is why modern sysadmins combine technical depth with operational discipline.
Clean data, smart AI
AI is only as good as the data it trains on. And clean, usable data does not happen by default.
Sysadmins are responsible for building and maintaining that data foundation. This includes databases, ingest pipelines, Extract, Transform, Load (ETL) processes, and storage systems.
Among the region’s most successful AI adopters, 40% credit high-quality data for their success. That falls squarely under the sysadmin’s remit. They manage access controls, ensure compliance with laws like GDPR or PDPA, and reduce data clutter through deduplication.
Poor data hygiene doesn’t just lead to weak insights. It undermines trust in the system. When AI gives erratic results, the issue often traces back to broken or biased data, and the fix starts at the infrastructure level.
AI governance starts at the system level
Today’s AI systems cannot just be fast or powerful. They also need to be transparent, traceable, and compliant.
That means building in observability. Every interaction, deployment, and compute cycle needs a clear audit trail. Sysadmins are the ones setting up those systems, maintaining logs, and keeping infrastructure accountable.
They are also responsible for what is called “model hygiene.” This includes removing outdated datasets, managing storage, and cleaning up broken workflows that affect performance or fairness. These tasks may sound routine, but they prevent real-world consequences.
As regulations evolve across Southeast Asia, organisations will need governance built into the infrastructure rather than bolted on after the fact. And when things go wrong, it is sysadmins who find the cause, restore the systems, and ensure it does not happen again.
The quiet force behind AI success
Data scientists drive innovation. Engineers bring ideas to life. But sysadmins make it all work.
They keep systems online, secure, and scalable. AI transformation is not just about advanced models; it also relies on the infrastructure supporting them. And at the heart of that infrastructure is a team of admins working behind the scenes.
AI has reshaped the role of system administrators. They are no longer just managing servers. They are enabling enterprise innovation. They reduce risk, support growth, and help organisations scale without losing control.
In Southeast Asia, countries like Singapore, Malaysia, and Indonesia are putting AI at the core of national digital strategies. As these plans move from vision to execution, sysadmins will play an even greater role in bringing them to life.
Their names might not appear in press releases or boardroom presentations. But every AI deployment that works—and works well—owes a lot to a sysadmin who kept it together.
Matthew Hardman is Chief Technology Officer, APAC, Hitachi Vantara