Case Study · Healthcare

Production IT infrastructure for Sofiamed Hospital

Since 2023 we manage the entire IT infrastructure of one of Bulgaria's large private hospitals — an environment where systems can't stop, because behind them are patients, doctors and care.

Client
Sofiamed Hospital
Sector
Healthcare
Environment
Multi-site, production
Partnership
Ongoing, since 2023
1,039
devices under monitoring, with full history
12+
managed servers (Windows and Linux)
since 2023
production environment under continuous management
24/7
monitoring of critical systems

Transparency note: we only state figures we can confirm. We don't publish invented statistics.

The context

A hospital runs around the clock. Reception, the lab, imaging, the pharmacy and the clinical departments all depend on systems being available — not "most of the time," but continuously. When we took on the engagement, the goal was clear: make the IT environment predictable, documented and resilient, without disrupting day-to-day operations.

The challenges

A real production environment comes with constraints that don't exist in a test lab:

  • Zero tolerance for downtime. Maintenance and changes must be planned so they don't affect patient care.
  • Many locations, one logic. Separate sites must operate as one secure network, with stable connectivity between them.
  • A heterogeneous environment. Windows and Linux servers, different applications and devices that must be managed consistently.
  • Sensitive data. Access, security and backups must meet high standards.
  • A large number of devices and users. Without order, requests get lost and assets become invisible.

Real responsibilities, every day

This isn't a one-off project. It's the ongoing operation of live infrastructure.

Windows and Linux servers

Configuration, updates, monitoring and maintenance of production servers.

Network administration

Managing the network, segmentation and stable connectivity between sites.

Site-to-site VPN

Securely connecting locations into one shared, encrypted network.

Monitoring and alerting

Watching availability and performance to catch problems early.

Backups and verification

Automated backups with periodic verification of real restores.

Incident response

A defined process for quickly isolating and restoring service.

End-user support

A helpdesk for staff — requests, priorities and traceability.

Asset management

Inventory of 1,039 devices with history and current state.

Documentation

Maintained documentation of network, servers, access and procedures.

Automation

Automating routine tasks to reduce manual work and errors.

Custom software

Development and maintenance of the helpdesk system described below.

Multi-site operations

Coordinated management of all sites as one environment.

A helpdesk system built around the real process

Off-the-shelf helpdesk products didn't match how the hospital works. So we built our own — for tickets, inventory and on-call scheduling — which we manage and evolve in real use.

helpdesk / dashboard
Helpdesk dashboard — overview of tickets and system status (anonymized data)
Ticket categories — classification of IT requests by type (anonymized data)
Asset inventory — hardware asset management (anonymized data)
AI Assistant — built-in AI for routine questions (anonymized data)

All screens use anonymized data. We don't show real names, patient data or sensitive information — even in our own case studies.

Our approach

We started the way we start every engagement — with an audit and mapping, to understand what exists and where the risks are. From there we built up systematically:

Stabilize and order

We bring backups to a state where they actually restore, arrange access along least-privilege, and document the network and servers so knowledge doesn't live in one person's head.

Connectivity between sites

Through site-to-site VPN the locations work as one secure network, and remote access is controlled and encrypted.

Visibility and response

Monitoring and alerting warn us of deviations before they become incidents. When a problem does occur, we follow a defined process for fast restoration and then document what happened.

Order in requests and assets

The custom helpdesk introduced traceability — requests no longer get lost in emails, and 1,039 devices have history and current state in one place.

Simplified, anonymized network schematic
SITE A SITE B MikroTik · Firewall network segmentation VLAN · workstations VLAN · servers MikroTik · Firewall network segmentation VLAN · workstations VLAN · servers SITE-TO-SITE VPN · encrypted tunnel Internet Remote access controlled · VPN

Illustrative schematic. Real addresses, names and segment counts are anonymized — exactly as we anonymize data in our own case studies.

What changed

Qualitative impact, described honestly — without attributing invented numbers.

Predictability

The infrastructure runs quietly and stably. Problems are caught early through monitoring instead of surprising anyone.

Transparency

A documented environment and traceable requests make it clear who does what and the state of everything.

Lower risk

Verified backups, controlled access and a segmented network reduce the risk of data loss and downtime.

We manage critical infrastructure. We can take on yours too.

We start with an audit and clear priorities — no obligation.