Skip to content

Engineering

Design, migration and automation — with ownership of the outcome

Network, datacenter, cloud and software engineering. Every project starts with an audit and ends with documentation and handover.

The problem

The project ended; ownership of the result never began

The integrator builds it, signs the acceptance note and leaves. What stays behind is a configuration nobody can explain, a diagram that was accurate a year ago, and a team that avoids changes because it cannot predict what a change will touch. The next project gets built on top of that layer.

  • Design decisions were never written down
  • Every change runs in production for the first time
  • The vendor is chosen by habit rather than by requirements
  • Handover means an archive, not knowledge

What we deliver

Design, migration and automation — with ownership of the outcome

Audit and current state

Topology, configurations, SPOFs, licences and technical debt in one prioritised document.

Target design

An architecture with alternatives and the cost of each — with a recommendation, not a single option.

Validated plan

Migration and rollback tested in the lab, then broken into work windows.

Implementation

In stages, each with its own acceptance criteria.

Documentation and runbooks

Diagrams, addressing plan and procedures in a format your team will keep updated.

Handover and training

Your engineers run the changes themselves; we stay on a retainer if you want one.

Architecture

Architecture

  1. 01AuditTopology, SPOFs and technical debt
  2. 02DesignAlternatives, and the cost of each
  3. 03ValidateMigration and rollback, in the lab
    • EVE-NG
    • Rollback test
  4. 04ImplementIn stages, each with acceptance criteria
  5. 05DocumentDiagrams, the addressing plan, runbooks
  6. 06HandoverYour team runs the changes themselves
How an engineering project runs — it starts with an audit and ends with a handover

Operator consoles

Operator consoles

These exact systems run on the group's own infrastructure — the screenshots are processed before publication.

Blurred Proxmox VE interface showing cluster nodes, containers and resource usage.

The virtualization cluster

A multi-node Proxmox cluster: KVM virtual machines and LXC containers, NFS/HA storage, a separate backup NAS and migrated ESXi hosts. This website runs on it too — in a container, on the same infrastructure we sell.

Proxmox VEKVM / LXCHA storageESXi migration
Blurred Proxmox Datacenter Manager — aggregate state of nodes, virtual machines, containers and backup servers.

8 clusters from one interface

Proxmox Datacenter Manager brings independent Proxmox VE clusters into one place. In our own infrastructure it manages 8 of them, spread across separate physical locations — 3 of those outside Georgia. The frame also shows the scale: 211 nodes online, 1,480 virtual machines and 127 containers — one view, one procedure.

Proxmoxmulti-clustermulti-sitePBS
Blurred vSphere Client — the virtual machine inventory and one machine's resource usage.

vSphere — the cluster, its hosts and the workloads on it

A vCenter with several ESXi hosts and the services placed on them: web, billing, a domain controller, backup agents. This is the environment where VMware and Proxmox coexist — part of it still on vSphere, part already migrated. One team operates both, under one backup policy.

vSphereESXivCentermigration
Blurred TrueNAS interface showing storage pools and disk health.

Storage — ZFS pools and replication

ZFS storage on a RAIDZ2 topology — 12 disks in one vdev, roughly 33 TiB usable, ECC memory with most of it serving as ZFS cache. And the part that matters: disks with errors, zero; scrub and scan, clean. This is the layer underneath the virtualization cluster and the backups.

TrueNASZFSRAIDZ2ECC
Heavily blurred billing system screen with a subscriber list — no personal data is legible.

ISP billing and subscriber management

The whole subscriber lifecycle in one system: tariffs, balances, services (internet, Wi-Fi, IPTV), statuses, SMS notifications, logs and financial reports — wired into provisioning. We built it, and it runs a real ISP.

in-housebillingprovisioningIPTV
The phpIPAM dashboard — aggregate statistics for subnets, VLANs and addresses, with usage charts.

IPAM — the address space in one system

A live phpIPAM deployment: subnet hierarchy, VLAN domains, VRFs, devices and locations in one searchable database. Only the aggregate statistics are public — customer names, individual subnets and locations are not on the frame, and should not be.

phpIPAMIPv4 / IPv6VLANVRF
Blurred EVE-NG topology showing routers, switches, peerings and management network links.

The digital twin — the lab that comes before production

A topology built from real network operating systems: border routers, IXP and CDN peerings, the aggregation layer, VLANs and the management network — production, replicated in the lab. Migration, failure and rollback run here first; only then does anyone touch the live network.

EVE-NGIOS XRBGPpre-production

Capabilities

Capabilities

Every item is marked: verified production experience, or engineering capability.

BGP and multi-homed edge

Proven
  • BGP
  • OSPF
  • RPKI

MPLS core and L3VPN

Capability
  • MPLS
  • MP-BGP
  • L3VPN

Virtualization clusters

Proven
  • KVM
  • Proxmox VE
  • VMware

EVPN-VXLAN fabric

Capability
  • Arista
  • Cumulus
  • FRR

Ceph / ZFS storage

Capability
  • Ceph
  • ZFS
  • NVMe

Automation and IaC

Capability
  • Ansible
  • Terraform
  • GitOps

Internal portals and APIs

Proven

A hosting control panel with hourly billing and provisioning automation — a working product, not a prototype.

  • Next.js
  • PostgreSQL
  • REST

Purpose-built operational tooling

Proven

When an off-the-shelf product does not answer the question, we build our own: a NOC console with live path monitoring, and the 33 diagnostic tools on noc.com.ge — both in daily use on our own network.

  • OpenPath
  • noc.com.ge
  • PHP / TypeScript

Process

Process

  1. 01

    Audit

  2. 02

    Design

  3. 03

    Validate

  4. 04

    Implement

  5. 05

    Document

  6. 06

    Handover

Technology stack

Technology stack

Routing and switching
CiscoJuniperMikroTikAristaFRR
Virtualization
VMware vSphereProxmox VEKVMLXC
Storage
TrueNASZFSNFSiSCSI
Automation
AnsibleTerraformcloud-initGit
Development
TypeScriptNext.jsPHPPostgreSQLREST API
Validation and monitoring
EVE-NGZabbixPrometheusGrafanaLibreNMS

Engagement model

Engagement model

Project

A one-off scope: audit, migration or implementation with a fixed outcome.

Retainer

Monthly engineering hours — specialist access on demand.

Use cases

Use cases

A migration against a deadline

A licence or hardware support contract is expiring. We produce the target design, prove it on a digital twin and move in stages — the deadline drives the schedule, not the architecture.

Two networks become one

After a merger the address space overlaps, VLAN numbering differs and each side has its own procedures. We produce one plan and join them without stopping the service.

The in-house team cannot keep up with growth

The engineers are consumed by day-to-day operations and never start the project. We take the project work while the team keeps production running, then hand it back with the documentation.

FAQ

FAQ

Will you tie us to one vendor?

No. We operate Cisco, Juniper, MikroTik, Arista, VMware and Proxmox in production — the recommendation comes after the audit and follows your existing hardware, budget and team experience.

Do you only design, or implement as well?

Both. An audit and a design can be bought on their own, but the standard project ends with implementation, documentation and handover — and afterwards, if you want it, with us operating the result.

Do you stay after the project?

Whichever you prefer: handover only, monthly engineering hours, a co-managed model, or fully managed operation. In every case the documentation is yours.

Tell us about your infrastructure