Cloud infrastructure & migration
Landing zones, infrastructure as code, migrations and cost control on Azure, Google Cloud and IBM Cloud.
DevOps consulting · Platform engineering · AI agents
DevOps, platform and AI engineering, secure from the first commit.
Integritasec is a DevOps and DevSecOps consultancy that designs, automates and runs platforms across cloud, on-premise and air-gapped environments, then puts AI to work on top of them: AI agents and automation, private AI on your own data, and AI infrastructure you control.
- Certified Ethical Hacker
- Azure Security Engineer
- ISO 27001 Foundation
One stack, built and run end to end
AI
- Internal AI on your data
- AI agents & automation
- AI infrastructure
- MLOps & LLMOps
Platform
- Kubernetes
- CI/CD
- Observability
- Security
Infrastructure
- Private cloud
- Public cloud
- On-premise
- Air-gapped
Runs inside your network. Your data stays yours.
Services
DevOps, platform and AI engineering services.
Take a single piece or the whole stack. Everything is delivered as code, documented, and handed over so your team can run it.
01
Platform engineering
Where your software runs and how it gets there.
On-premise & air-gapped infrastructure
VMware, Hyper-V and bare-metal platforms, including fully disconnected environments with no internet access.
Kubernetes consulting
Cluster design, hardening, upgrades and day-two operations. Upstream Kubernetes and OKD / OpenShift.
CI/CD pipelines & automation
Pipelines on GitLab CI, Jenkins or TeamCity that build, test, scan and release without manual steps.
02
Cloud security & observability
Knowing what is happening, and keeping it safe.
Monitoring & observability
Metrics, logs and alerting with Prometheus, Grafana, Loki, Elastic and Zabbix. Alerts your team trusts instead of mutes.
DevSecOps & cloud security
Security as code: scanning and policy gates inside the pipeline, hardened clusters and networks, and practices aligned with ISO 27001.
03
AI engineering
AI that works in production, not only in a demo.
AIOps
AI applied to operations: alert correlation, anomaly detection, faster incident triage and automated runbooks.
AI infrastructure, MLOps & LLMOps
The platform under your models: GPU capacity and model serving, self-hosted or through cloud APIs, with evaluation, guardrails and cost monitoring.
AI agents & automation
Agentic AI and AI orchestration: agents and automated workflows connected to your tools and systems, with approvals, audit logs and spending limits.
Generative AI consulting
Getting your teams using AI safely: internal assistants, coding assistants, and clear rules for what data goes where.
Private cloud + enterprise AI
Private AI on your own cloud, with an assistant that knows your business.
Most AI rollouts assume everything can be sent to a public service. Regulated and security-sensitive companies cannot work that way.
Integritasec builds the hosting and the AI together, so your company gets an assistant that answers from your own documents, code and systems, inside infrastructure you control.
More on private AI and on-premise AI
Private cloud hosting
An internal cloud on your hardware or in a dedicated tenancy: Kubernetes, storage, networking, backups and monitoring, built and operated for you.
Internal AI on your data
An AI assistant grounded in your knowledge base, tickets, wikis and repositories through retrieval-augmented generation (RAG). It respects existing access rights, and your data is not used to train public models.
The model that fits
Claude and other leading models through your own cloud account, or open models self-hosted on your GPUs when nothing may leave the network.
Engagements
Ways to work together.
Start here
Assessment
A short, fixed-scope review of your infrastructure, pipelines, security posture and AI readiness.
- Written findings, ranked by risk and effort
- A concrete plan your team can act on
- No commitment to further work
Defined outcome
Project
One clear deliverable, scoped and priced up front.
- Cloud or on-premise migration
- Kubernetes platform or pipeline rebuild
- Private cloud with an internal AI assistant
Continuous
Ongoing partnership
A set number of days each month for the platform work your team never gets to.
- Upgrades, hardening and reviews
- Pipeline and tooling improvements
- A senior engineer on call for advice
Approach
From first call to handover.
-
1
Call
Thirty minutes on what you run today, what hurts, and what you need it to do.
-
2
Assess
A look at the real systems, then a written scope with outcomes, timeline and price.
-
3
Build
Work delivered in small, reviewable steps, in your repositories, with security built in.
-
4
Hand over
Documentation, runbooks and a walkthrough, so your team owns what was built.
About
You work with the engineer doing the work.
Integritasec was founded in 2023 in Ljubljana by Filip Obradović, a DevSecOps engineer and DevOps team lead with ten years in infrastructure, from system engineering in the field to running Kubernetes platforms in the cloud, on-premise and in air-gapped environments.
No account managers and no hand-off to juniors. The person on the first call is the person who designs, builds and documents your platform.
- 10 yrs
- in infrastructure and operations
- 2023
- Integritasec founded
- 3
- working languages: English, Slovenian, Serbian
Certifications
- Certified Ethical Hacker (CEH)
- Microsoft Certified: Azure Security Engineer Associate
- ISO 27001 Foundation
- ISO 9001 Foundation
Tools in daily use
- Kubernetes
- OKD
- Docker
- GitLab CI
- Jenkins
- TeamCity
- Azure
- GCP
- IBM Cloud
- VMware
- Hyper-V
- Prometheus
- Grafana
- Loki
- Elastic
- Zabbix
- Kafka
- PostgreSQL
- ClickHouse
- NGINX
- HAProxy
Questions
Asked before the first call.
What does a DevOps consultant do?
A DevOps consultant shortens the path from code to production and keeps it reliable. In practice that means cloud or on-premise infrastructure, Kubernetes, CI/CD pipelines, monitoring and security, built as code and handed over to your team.
Can AI run on-premise without sending data to the cloud?
Yes. Open language models can be self-hosted on your own GPUs, including in air-gapped networks, so prompts and documents never leave your infrastructure. Where a cloud service is acceptable, models such as Claude can run through your own cloud account under your own data controls.
What is the difference between AIOps and LLMOps?
AIOps uses AI to run IT operations better: fewer noisy alerts, faster incident triage, automated runbooks. LLMOps is the practice of deploying and operating large language models themselves: serving, evaluation, guardrails, cost and performance monitoring.
How does an internal AI assistant learn our company data?
Usually through retrieval-augmented generation (RAG): your documents, tickets and code are indexed inside your environment, and the assistant looks up the relevant pieces when it answers. Access rights are respected, and fine-tuning is added only where it clearly pays off.
Do you work with companies outside Slovenia?
Yes. Work is remote-first in English, Slovenian or Serbian, with on-site visits in Slovenia when a project needs them.
How do we get started?
Book a 30-minute call. If there is a fit, the usual first step is a short assessment that ends with written findings and a concrete plan.
Book a call
Pick a time. Thirty minutes, no slides.
Bring the problem as it is. You will leave with a clear view of what it would take to fix, whether or not we work together.
Online scheduling is being set up. Until then, send a short message and you will get a reply with available times.