CyberLab Solutions

CyberLab Solutions We are a DevOps and Software Development company built on trust, technology, and collaboration. Let’s build something lasting, together.

Defined by team spirit, ownership, and a growth mindset, we move as one with our engineers and partners.

Ljubljana, 22.10. See you there. 🇸🇮It’s our first time at AWS Community Day Adria, the community conference built by six...
14/10/2026

Ljubljana, 22.10. See you there. 🇸🇮

It’s our first time at AWS Community Day Adria, the community conference built by six AWS user groups from across the region.

Where to find us:
▪️11:30, Fabiani: Andrija on “One Firewall to Rule Them All: Hub & Spoke on AWS”
▪️All day, expo area: the CYBERLAB booth

Come say hi and tell us what you’re building.

Thank you to the organizers and volunteers who make this happen on their own time. 🙌

Talk details: link in bio.

Vidimo se v Ljubljani!👋

cyberlab

Last week, Aleksa Kralj walked us through the Model Context Protocol (MCP), the standard that lets any AI app connect to...
13/10/2026

Last week, Aleksa Kralj walked us through the Model Context Protocol (MCP), the standard that lets any AI app connect to tools and data without custom integrations. 🔌

Going beyond the basics, he broke down the latest spec: MCP is now stateless, and several features have been deprecated.

Our key takeaway: your AI trusts every MCP server it connects to, so vet each one like any other dependency.

Thank you, Aleksa! 🙏

First time at Community Day Adria. 🇸🇮On 22 October, our team is heading to Ljubljana for the region’s biggest Amazon We...
09/10/2026

First time at Community Day Adria. 🇸🇮

On 22 October, our team is heading to Ljubljana for the region’s biggest Amazon Web Services (AWS) community conference, built by six AWS user groups from across the Adria region.

We’re coming to learn, to meet the people who build and run things on AWS in our region, and to share some of what we’ve learned along the way.

That includes “One Firewall to Rule Them All: Hub & Spoke on AWS”, a talk by our Lead DevOps Engineer Andrija Vojnovic at 11:30 on how to design an AWS network that grows with you, not against you.

The rest of the day, you’ll find us at our booth in the expo area. Come by, say hi, and bring your AWS questions.

The event is officially sold out, but the organizers just made room for 50 more people, and that’s the hard limit. If you’ve been on the fence, this is the last chance. Grab your ticket via the link in our bio.

Vidimo se v Ljubljani! 👋

True scalability in healthcare platforms is less about infrastructure growth and more about eliminating operational frag...
07/10/2026

True scalability in healthcare platforms is less about infrastructure growth and more about eliminating operational fragmentation. 🏥

In this engagement, we modernized a Med-Tech platform struggling with fragmented integrations, manual infrastructure processes, and inconsistent device event handling as adoption rapidly increased.

The core challenge wasn’t device scale; it was operational complexity across cloud environments, APIs, deployments, and compliance-sensitive workflows.

We addressed this with Infrastructure-as-Code (Terraform, Terragrunt), event-driven ingestion pipelines, centralized API governance, automated CI/CD delivery, and unified observability.

Key outcomes:
▪️ Standardized, reproducible infrastructure across environments
▪️ Scalable event-driven IoT ingestion architecture
▪️ Centralized API governance with enforced security policies
▪️ Automated CI/CD with approvals and rollback
▪️ Unified observability (logs, metrics, tracing)
▪️ Platform ready for global and compliant expansion

The result is a healthcare platform built for predictable scale, operational resilience, and continuous delivery, without disrupting live clinical operations.

Modernization succeeds when engineering teams reduce operational friction, not just add new technology.

Read the full case study. Click on the link in our bio 📲

PlatformEngineering InfrastructureAsCode

As part of the prep for our internal hackathon, our colleague Kristina, a software developer, walked the team through Cl...
24/09/2026

As part of the prep for our internal hackathon, our colleague Kristina, a software developer, walked the team through Claude Cowork - a way of working with AI that doesn't just answer questions but takes on the task itself.

A chatbot tells you how to do something. Cowork actually does it: it reads your files, connects to the tools you already use (Gmail, Slack, Drive, calendar), and saves a finished document back into your folder.

What caught our attention: of 1.2 million Cowork sessions Anthropic analyzed, 33% went to business process and operations work, while software development and DevOps together made up under 16%. It's not a coding tool - 𝗶𝘁'𝘀 𝗳𝗼𝗿 𝗲𝘃𝗲𝗿𝘆𝘁𝗵𝗶𝗻𝗴 𝗮𝗿𝗼𝘂𝗻𝗱 𝘁𝗵𝗲 𝗰𝗼𝗱𝗲, like turning a week of tickets into one status update or pulling a monthly report from raw spreadsheets.

We also covered how to set it up your way (instructions, skills, connectors, plugins) and how to use it safely: a dedicated folder it can't step outside of, prompts that spell out exactly what an action means, and reading its plan while the task runs, not after it's done.

Thanks, Kristina, for sharing these useful insights!

Branislav Gvozdenov has successfully completed an 8-week hands-on DevOps internship focused on real infrastructure and m...
22/09/2026

Branislav Gvozdenov has successfully completed an 8-week hands-on DevOps internship focused on real infrastructure and modern cloud practices, guided by his mentor, our colleague Ivan Inđić, DevOps Engineer. 💡

Over the course of the program, he:
▪️Deployed a vanilla Kubernetes cluster on AWS EC2 using kubeadm, building a solid understanding of cluster architecture and its core components from the ground up.
▪️Used Helm to deploy the tools a production environment actually needs - DNS management, HTTPS certificate issuance and renewal, monitoring, and centralized logging.
▪️Implemented GitOps with ArgoCD, gaining hands-on experience with declarative infrastructure management.
▪️Set up a CI pipeline with GitHub Actions, closing the loop from code change to deployed infrastructure.

What made this valuable was the sequence: each stage built the context needed for the next. By the end, the concepts weren’t abstract, they were something he had actually built, configured, and operated.

Congratulations, Branislav, and thank you for your dedication and active participation throughout the internship, we’re excited to welcome you to the CYBERLAB team as a full-time employee! 🙌

Branislav Gvozdenov has successfully completed an 8-week hands-on DevOps internship focused on real infrastructure and m...
22/09/2026

Branislav Gvozdenov has successfully completed an 8-week hands-on DevOps internship focused on real infrastructure and modern cloud practices, guided by his mentor, our colleague Ivan Inđić, DevOps Engineer.

Over the course of the program, he:
▪️Deployed a vanilla Kubernetes cluster on AWS EC2 using kubeadm, building a solid understanding of cluster architecture and its core components from the ground up.
▪️Used Helm to deploy the tools a production environment actually needs - DNS management, HTTPS certificate issuance and renewal, monitoring, and centralized logging.
▪️Implemented GitOps with ArgoCD, gaining hands-on experience with declarative infrastructure management.
▪️Set up a CI pipeline with GitHub Actions, closing the loop from code change to deployed infrastructure.

What made this valuable was the sequence: each stage built the context needed for the next. By the end, the concepts weren't abstract, they were something he had actually built, configured, and operated.

Congratulations, Branislav, and thank you for your dedication and active participation throughout the internship, we're excited to welcome you to the CYBERLAB team as a full-time employee! 🙌

Most infrastructure friction isn’t a dramatic outage. It’s mundane: updating one agent shouldn’t require touching thirty...
17/09/2026

Most infrastructure friction isn’t a dramatic outage. It’s mundane: updating one agent shouldn’t require touching thirty task definitions and pinging five teams.⚙️

That’s coupling. It stays invisible until the moment you need to move fast.

The real story behind this ECS update isn’t the feature. It’s the principle: operational concerns and application code have different owners, different update cadences, and never should have shared a lifecycle in the first place.

Details in the carousel.

Link to the blog in the bio. ⬆️

Most infrastructure friction isn't a dramatic outage. It's mundane: updating one agent shouldn't require touching thirty...
17/09/2026

Most infrastructure friction isn't a dramatic outage. It's mundane: updating one agent shouldn't require touching thirty task definitions and pinging five teams.

That's coupling. It stays invisible until the moment you need to move fast.

The real story behind this ECS update isn't the feature. It's the principle: operational concerns and application code have different owners, different update cadences, and never should have shared a lifecycle in the first place.

Details in the carousel.

Link to the blog in the comments.

We’re proud to share that our Lead DevOps Engineer, Andrija Vojnović, is speaking at AWS Community Adria in Ljubljana.Hi...
14/09/2026

We’re proud to share that our Lead DevOps Engineer, Andrija Vojnović, is speaking at AWS Community Adria in Ljubljana.

His talk, “One Firewall to Rule Them All: Hub & Spoke on AWS,” covers centralizing network inspection across multi-VPC AWS environments using Cloud WAN as the backbone.

See you on 22 October in Ljubljana!

More info about his talk via link in bio. ☝️

Address

Cara Dušana 2
Novi Sad
21000

Opening Hours

Monday 09:00 - 17:00
Wednesday 09:00 - 17:00
Friday 09:00 - 17:00

Website

Alerts

Be the first to know and let us send you an email when CyberLab Solutions posts news and promotions. Your email address will not be used for any other purpose, and you can unsubscribe at any time.

Shortcuts

Share

Category