Kubernetes
Production-grade clusters, hardened workloads and dependable day-two operations. We design greenfield platforms, lead migrations and stabilise environments that have become fragile.
RKE2, GKE, EKS, Istio, Helm, Argo CD
DevOps consultancy
DevOps engineering for infrastructure that has to hold.
Decoding Labs designs, builds and operates the platforms enterprises depend on. Kubernetes, GitLab, Linux, Azure DevOps and CI/CD, delivered by senior engineers into insurance, banking, energy and payments environments for enterprises worldwide.
12+
Years of infrastructure and DevOps engineering across insurance, banking, energy, payments and telco.
Currently delivering
Swiss reinsurance
Smart-metering technology
An analytics platform for around 350 users at a reinsurer, and the CI/CD and Kubernetes platform behind a smart-metering manufacturer's R&D.
300+
Repositories consolidated and migrated to GitLab SaaS for an R&D organisation, with every merged change deploying automatically.
About Decoding Labs
A specialist DevOps consultancy for regulated enterprises.
Decoding Labs builds, modernises and operates the infrastructure enterprises depend on. Our engineers run platforms today for a Swiss reinsurer and a smart-metering manufacturer, and have done so for banks, an insurance group's IT provider and a card-payment processor: environments where every change is audited and every outage is expensive.
The practice is deliberately narrow: Kubernetes, GitLab, Linux, Azure DevOps, CI/CD, infrastructure automation and platform engineering, across Azure, AWS, GCP, Swisscom Cloud and private cloud. Depth over breadth, delivered by engineers who have run these systems in production and on call.
The standard is simple. Reliable systems, clean automation, predictable operations and decisions that hold up in front of a security review.
- KubernetesRKE2, EKS and GKE in production. GitOps with Argo CD, Helm and Istio.
- Azure DevOpsMicrosoft certified Azure DevOps Engineer Expert. Delivery tooling for a Swiss reinsurer.
- TerraformHashiCorp certified. Terraform, Ansible and Puppet across mixed Windows and Linux estates.
- SwitzerlandSix years across two engagements at a Swiss reinsurer, on Swisscom Cloud.
Services
Practical engineering for the platforms enterprises run.
Platforms
Linux infrastructure
Patched, observable and predictable Linux estates, built and operated to enterprise standards. Infrastructure that does not surprise anyone on a Sunday night.
Ubuntu, RHEL, Puppet, Ansible
Cloud and platform engineering
Internal platforms that reduce cognitive load for engineering teams. Self-service where it pays off, guardrails where they have to hold.
Azure, AWS, GCP, Swisscom Cloud, vSphere
Delivery
GitLab and CI/CD
GitLab set up as a complete delivery platform. Runners, pipelines, security scanning, release automation and a developer experience that scales with the organisation.
GitLab CI/CD components, runners, security scanning, Renovate
Azure DevOps
Boards, repos, pipelines, environments and policies wired into a delivery model your teams actually use. Microsoft-certified, opinionated and fast.
Pipelines, repositories, branch policies
Infrastructure automation
Terraform, Ansible and GitOps. Reproducible environments, codified change and a clean path from commit to production.
Terraform, Ansible, Puppet, Argo CD ApplicationSets
Operations
DevOps audit
A written, independent assessment of your CI/CD, infrastructure and operations, with clear findings and a prioritised action plan.
Fixed scope, written report
Modernisation
Replatforming, pipeline rewrites and Kubernetes adoption. We turn fragile setups into calm, maintainable platforms.
Legacy CI to GitLab or GitHub Actions, artifact re-platforming
Reliability and operations
SLOs that mean something, sustainable on-call and observability that answers the question before it is asked.
Prometheus, Grafana, OpenSearch, Kibana alerting
The tools we run in production.
- Linux
- Kubernetes
- GitLab
- Azure DevOps
- GitHub Actions
- Terraform
- Ansible
- Puppet
- Helm
- Argo CD
- Istio
- JFrog Artifactory
- RabbitMQ
- Vault
- Prometheus
- Grafana
- OpenSearch
- Azure
- AWS
- GCP
- Swisscom Cloud
The shape of a delivery platform we build.
One reference layout. The tools change from client to client, the principles do not: everything as code, one path from merge request to production, secrets and observability built in from the first day.
Selected work
Platforms we have built and run.
Reinsurance, Switzerland
Platform engineering for a Swiss reinsurer on Swisscom Cloud
Two engagements with the same client. Four and a half years running infrastructure and CI/CD for the group's tax IT: Terraform provisioning and Ansible deployments across a mixed Windows and Linux estate, Elasticsearch clusters behind an application critical to quarterly closings, later replaced with OpenSearch to gain built-in authentication and cut a licence cost, and two server migrations that earlier attempts had failed. Since 2025, running an analytics workbench platform for around 350 users, with delivery tooling on Azure DevOps and provisioning with Puppet, kept stable through nine months of unstable vendor releases until the root cause was traced to the vendor's own release.
- Engagement
- Two engagements, 2019 to 2023 and since 2025
- Environment
- Regulated, design-governance reviews with security
- Scope
- Infrastructure, CI/CD, platform operations, migrations
- Stack
- Swisscom Cloud, Azure DevOps, Terraform, Ansible, Puppet, OpenSearch
Latest client review: 10 out of 10, citing end-to-end ownership across a wide task scope.
Energy technology, Switzerland
GitLab SaaS migration and a GitOps Kubernetes platform for a smart-metering manufacturer
A self-hosted GitLab estate of more than 300 repositories with fragmented pipelines, consolidated and migrated to GitLab SaaS through a central library of versioned CI/CD components: one source of truth for every pipeline, used by around 20 developers, with every merged change deploying automatically and security scanning mandatory. Infrastructure definitions brought into a single GitOps repository with Argo CD ApplicationSets. We run the R&D Kubernetes platform: RKE2 with Istio, Traefik and External Secrets Operator, 16 RabbitMQ instances moved to the cluster operator, artifacts re-platformed onto JFrog Artifactory with Renovate and OSV alerts, and a build-versus-cloud assessment for GCP with a multi-year cost comparison.
- Engagement
- Ongoing since 2023
- Environment
- R&D and delivery organisation, product moving to SaaS
- Scope
- CI/CD platform, GitOps, Kubernetes, artifact management, capacity planning
- Stack
- GitLab CI/CD, RKE2, Istio, Argo CD, Helm, JFrog Artifactory, GKE
Client review: rated High Impact, described as critical for deliveries and driving key initiatives.
Earlier work
- Banking. Assessment of Windows containers on AKS and OpenShift for a bank's legacy .NET and Java monoliths, and migration of its TeamCity pipelines to GitHub Actions on self-hosted Windows runners.
- Insurance group IT. CI/CD modernisation for a legacy Java application on WebSphere and IBM MQ: Jenkins pipelines, Docker images, Helm charts and secrets handling, plus a custom search engine hosted on AWS EKS.
- Banking operations. Transition of a German bank's connectivity operation to a new team in Czechia: know-how captured, documentation written, the new team hired and the handover completed.
- Payments. Around-the-clock operation of a card-payment switch serving banks in three countries, with disaster-recovery exercises and production deployments during the Apple Pay and Google Pay launches.
Client names and references are available on request.
How we work
From the first call to a platform your team can run.
Every engagement follows the same four steps, whether it is a two-week audit or a multi-year retainer. Each one starts with a thorough understanding of the current state and ends with documentation your team can operate from.
-
Assess
A hands-on review of your pipelines, clusters and operations. What is fragile, what is fine and what is quietly expensive.
-
Plan
A written roadmap with scope, sequencing and risks. Decisions made with the next three years in mind.
-
Build
Clusters, pipelines and automation delivered as code, with reviews that hold up in front of security and compliance.
-
Operate
Structured handover with documentation and runbooks, or an ongoing retainer with SLOs and sustainable on-call.
- Senior by default. The people who scope the work are the people who deliver it. Nobody learns on your platform.
- Built for the audit. Every change is codified, reviewable and defensible. Regulated environments are where we work every day.
- Plain answers. If a tool, pattern or migration is the wrong call, you hear it early. We would rather lose a project than recommend the wrong architecture.
Leadership
Ivan Servacek
Founder and principal engineer
Ivan founded Decoding Labs after more than twelve years in infrastructure and DevOps engineering across insurance, banking, energy, payments and telco. He leads the practice and stays hands-on: as the cloud engineer behind an analytics platform for around 350 users at a Swiss reinsurer, and as the engineer who runs the CI/CD and Kubernetes platform for a smart-metering manufacturer's R&D, where he consolidated a 300-plus repository estate and led its migration to GitLab SaaS.
Certified Microsoft Azure DevOps Engineer Expert and Azure Administrator, HashiCorp Terraform Associate, AWS Cloud Practitioner, Red Hat system administration and ITIL Foundation. Based in Slovakia, working with clients worldwide, remotely and on site.
Questions
What clients ask before the first call.
How does an engagement start?
With a short call to understand your platform and what needs to change, followed by a written proposal with scope and sequencing. Nothing is built before it is understood.
Which engagement models do you offer?
Fixed-scope audits, project delivery and ongoing retainers. All three are quoted in writing before the work starts.
Do you work on-site?
Most of the work happens remotely, inside your environment and your tooling, for clients anywhere in the world. On-site presence is arranged where the work requires it.
Can you work in regulated environments?
Yes. Our engineers have worked inside a Swiss reinsurer's design-governance process, a card-payment switch with around-the-clock on-call, and banks' change-controlled estates. Work is delivered as code, reviewed and documented so it holds up in front of security and compliance.
What happens when the project ends?
Every project ends with documentation and runbooks your team can operate from. Where a retainer makes sense, we stay on for operations, SLOs and on-call.
How do I get a quote?
Write to hello@decoding-labs.com with a paragraph about your platform. You will hear back within 24 hours on business days.
Contact
Let's talk about your platform.
hello@decoding-labs.comTell us about your infrastructure and what needs to change. You will hear back within 24 hours on business days and, where it makes sense, we arrange a call with the engineers who would do the work.
- Company
- Decoding Labs s. r. o.
- Address
- Sitnianska 6597/13
974 11 Banská Bystrica
Slovakia - Company ID
- 54976642
- Tax ID
- 2121832669
- VAT ID
- SK2121832669
- IBAN
- SK32 0900 0000 0051 9630 7081