By: Olena Midzhak

In healthcare technology, a failed deployment is often not just a technical inconvenience. When the systems behind clinical trials, patient data, and regulatory reporting go down, the consequences can extend far beyond a status page. It is an environment where infrastructure reliability is not simply a performance metric; it is a matter of trust.

That reality defines the work of Danylo Mikula, a DevOps and Infrastructure Architect at Medidata Solutions, a healthcare technology company that provides SaaS and data analytics platforms supporting clinical research worldwide. Over the past two years, Danylo has led a transformation that has significantly changed how the company’s platform engineering team builds, delivers, and maintains its infrastructure, turning what was once a slow, manual, and error-prone process into a system designed around predictability and attempted proof.

When Infrastructure Runs on Memory, Not Systems

When Danylo joined Medidata’s platform engineering team in late 2023, he found a group of skilled engineers managing a complex hybrid environment, multiple Kubernetes clusters spanning on-premises data centers and cloud infrastructure, along with thousands of virtual machines powering critical healthcare workloads. The technical talent was present, but the processes had developed organically over time and hadn’t kept pace with the growing scale of the operation.

Getting a new production environment ready could take weeks of coordinated effort across multiple teams. Configurations were distributed across the organization rather than managed from a central system, and operational knowledge naturally shifted as teams evolved. Updates to infrastructure components were frequently postponed because no one could be entirely sure what was running where.

The system worked, but it worked because people had to compensate for its gaps. That distinction mattered to Danylo.

“When reliability depends on someone remembering the right steps in the right order, it isn’t necessarily reliability. It’s luck that hasn’t run out yet.” – Danylo Mikula

Designing for Proof, Not Just Performance

Rather than layering new tools onto existing workflows, Danylo took a different approach. He restructured the way change enters the system. Every configuration, every deployment, every secret was moved into version-controlled repositories that serve as a single source of truth. Automated processes work to continuously ensure that what is defined in those repositories matches what is actually running in production.

The principle behind the shift, known in the industry as GitOps, is straightforward: if every change is recorded, reviewed, and applied through a consistent pipeline, then every deployment becomes traceable, and every environment becomes reproducible. For a healthcare technology company operating under strict regulatory requirements, traceability is not a mere convenience; it is a compliance necessity.

The results were significantly improved. Provisioning a production-ready environment went from a process that could take weeks to one that completes in approximately thirty minutes. But for Danylo, speed was a side effect. The real outcome was something harder to measure and more valuable to maintain: operational clarity.

“The goal was never to go faster for the sake of speed. It was to reach a point where we could answer a simple question at any moment: what exactly is running in production, and when did it change?” – Danylo Mikula

Photo Courtesy: Danylo Mikula

People, Not Just Pipelines

Colleagues who worked alongside Danylo during the transformation emphasize that the technical architecture was only part of the story. The harder work was cultural: helping experienced engineers adopt an entirely new way of thinking about infrastructure delivery.

“What set this work apart was the patience behind it. Danylo didn’t just build a system and hand it over. He brought people along, explained the reasoning, and made sure the team felt ownership, not just compliance.” – Monik Gandhi, Director, Cloud Engineering, Medidata Solutions

That human-centered approach was critical. Engineers who had spent years working with manual, imperative processes were gradually introduced to a declarative model where the system’s desired state is defined once and maintained automatically. Rather than mandating the change from above, Danylo invested time in education and demonstration, running workshops and walkthroughs that focused not just on how the new system worked, but on why it was designed the way it was.

“The biggest change wasn’t in the tools, it was in confidence. Engineers who used to hesitate before making infrastructure changes now run routine updates without second-guessing, because the process is transparent and reversible.” – Labhesh Potdar, Manager, Cloud Engineering, Medidata Solutions

When Compliance Becomes a Byproduct

In healthcare technology, regulatory requirements are not optional, and demonstrating compliance requires more than good intentions. It requires evidence, records of what changed, when, and by whom.

Before the transformation, producing that evidence was a labor-intensive process. Configuration histories were incomplete, and audit trails had to be reconstructed manually. After the shift, compliance documentation became a natural byproduct of everyday operations. Every infrastructure change flows through a version-controlled pipeline, creating an automatic, tamper-resistant record.

Secrets management, the handling of passwords, certificates, and access credentials, was similarly centralized and automated, ensuring consistent rotation and access controls across the entire infrastructure. For a company whose platforms support clinical research and handle sensitive healthcare data, this shift represented a material improvement in security posture.

From Practice to Research

What distinguishes Danylo from many practitioners in the field is his commitment to documenting and sharing what he learns. In addition to his day-to-day work at Medidata, he has contributed research on infrastructure transformation strategies and common implementation failures to international scientific conferences.

His recent academic work includes a systematic taxonomy of anti-patterns, recurring mistakes that organizations make when adopting modern infrastructure practices, along with evidence-based mitigation strategies. The research quantifies how specific failures impact operational metrics like recovery time and change failure rates, providing other engineering leaders with actionable data rather than anecdotal advice.

“Every team makes similar mistakes when adopting these practices. If we can document those patterns and share them, we might save others from repeating the same costly lessons.” – Danylo Mikula

Looking Ahead

With the infrastructure delivery model now established, Danylo’s focus is shifting toward the next frontier: making reliability itself measurable. His current work involves building observability frameworks that define clear indicators of system health and trigger automated responses when those indicators degrade, moving the organization from reactive incident management to proactive governance.

It is a natural extension of the philosophy that has guided his career: that reliability should not be a matter of opinion or intuition, but an objective, demonstrable quality of the systems we build.

“The ideal infrastructure is the kind nobody has to think about. Not because it doesn’t matter, but because it was built well enough that it just works, and when it doesn’t, you can see exactly why.” – Danylo Mikula

Danylo Mikula is a DevOps and Infrastructure Architect at Medidata Solutions with over ten years of experience delivering cloud and platform engineering solutions in regulated industries. His work focuses on translating DevOps principles into measurable, repeatable reliability practices, emphasizing declarative workflows, infrastructure as code, and observability-driven governance. He has contributed research on GitOps adoption patterns and enterprise transformation strategies to international scientific conferences. More about his projects and technical work can be found on his personal website, showcasing his hands-on experience and product development approach.