HYBRID INFRASTRUCTURE ENGINEER

Eight years building and running infrastructure.

I'm Mateo, a systems engineer. Most of my career has been on-premise VMware: I built a datacenter from scratch, from the VLANs and the storage array up to vCenter and backups, and I've migrated more than 100 customers between platforms. These days I also work with Linux, Azure, AWS and Proxmox, and that's where I want to grow: hybrid infrastructure, built on what I learned in the datacenter.

DOWNLOAD CV ↓LINKEDIN PROFILE ↗

SANTIAGO DE COMPOSTELA · GALICIA, SPAIN8+ YEARS IN SYSTEMSVMware VCP-DCV · 2023

console · try a command

Try a command status · deploy · whoami · help

Type status for the live homelab status, or help for the rest.

…65648f1

BY THE NUMBERS

100+customersonboarded, run or migrated at cloud.gal

~1,000virtual machinesrunning 24/7 at cloud.gal

~3,000virtual machinesin DXC's main customer environment

3Pure Storage arraysrun from initial setup to vendor support cases

EXPERTISE

What I work with.

Most of my hours are in virtualization, storage and backup. I use the rest every day, but not as deeply, and I say so in each area.

VMware · ESXi · vCenter · Cloud Director

Virtualization & datacenter

This is where most of my experience is. I've installed, configured, upgraded and migrated VMware platforms, starting with the compatibility matrix and server firmware and finishing with customers running on top.

ESXi and vCenter 6.5, 7.x and 8.x
Clusters, HA, DRS and VDS
BIOS, firmware, drivers and custom ESXi images
Cross-vCenter migrations and customers on Cloud Director
Patching, version upgrades and compatibility matrices
Some NSX-T, vRealize and Aria, not in depth

VMwarevSphereESXivCenterCloud Director

Pick an area to see the details.

EXPERIENCE

Where I've worked.

At cloud.gal I owned projects from start to finish: from the first customer meeting to the follow-ups after go-live, with the design, build, migration and backups in between. Before that, at DXC, I ran VMware clusters for public-sector customers. Since 2026 I have been at Enxenio, focusing more on Linux and cloud.

MAR 2026 — PRESENT

EnxenioA Coruña · Hybrid

Infrastructure & Systems Engineer

The job here is mostly operations. I run Linux servers across production, staging and development, deploy to AWS and also work with Azure. I build and document VMs on Proxmox, set up Restic backups and snapshots, and automate deployments and routine tasks. Lots of log reading: Docker, systemd, Jenkins, GitLab, PostgreSQL. On one project I changed the configuration of a third-party VMware environment to fit what our application needed, acting as the go-between for the customer and our developers.

CURRENT

JUL 2024 — MAR 2026

cloud.galSantiago de Compostela · Hybrid

Virtualization & Backup Lead

I ran projects from start to finish: customer kick-off, requirements, reviewing their existing setup, deployment, hand-holding, backups, tuning and follow-ups. I designed and built the new VMware 8 datacenter and migrated customers into it from the old one. I also handled ESXi upgrades and patching, migrations between vCenters and arrays, from Azure to VMware and onto Cloud Director, and the Veeam and DR projects.

FEB 2023 — JUN 2024

DXC TechnologySantiago de Compostela · Remote

VMware Systems Administrator

I ran VMware clusters for AMTEGA and CIXTEC: installing and configuring new clusters on Lenovo, Dell and HP servers, lifecycle, patching and VM provisioning. I prepared compatibility matrices, ran and adapted PowerShell scripts, and used vRealize Orchestrator and Terraform. The main customer had about 600 servers and 3,000 virtual machines.

MAR 2021 — FEB 2023

Landra SistemasNarón

Level 2 Systems Technician · Roche

Level 2 support in hospital environments on Windows and Linux. I reviewed logs, fixed incidents and integrated healthcare applications over HL7. I automated tasks with Python and worked daily with clinical staff and vendors.

JUL 2020 — MAR 2021

Evelb TécnicasMilladoiro

Support Lead

Application deployments with Git and Jenkins, quality control and infrastructure operations on AWS and GCP.

AUG 2019 — JUL 2020

Plexus TechSantiago de Compostela

Project Coordinator

I coordinated teams of 10 to 15 technicians and national and international support services: planning interventions, tracking deadlines and handling escalations.

JUL 2017 — JUN 2019

Fermax ElectrónicaGalicia

Technical Lead · Galicia

Technical lead for Galicia: customer support, on-site and remote commissioning, troubleshooting and training.

HOMELAB

What I run at home.

Two mini PCs running Debian 13. One handles DNS and DHCP with AdGuard Home and monitors everything with Prometheus, Grafana and Uptime Kuma. The other runs the containers, including this website. I use it to try things I can't break at work. The site reaches the Internet through Cloudflare Tunnel, with no ports open on the router, and it redeploys itself every time I push a change to GitHub. You can see the live service status at mcouso.es/estado.

LOGICAL DIAGRAM · HOME NETWORK

INTERNET
GATEWAY / ROUTERMOVISTAR HGU
SWITCHINGGIGABYTE SWITCH
- ROUTER / WIRELESSXIAOMI AX3600Wi-Fi · OpenWrtSITE-TO-SITE VPN → REMOTE LAN
- LENOVO THINKCENTRE M715qHOMELAB-REDAdGuard Home · DNS + DHCPPrometheus · Grafana · exporters · Uptime Kuma
- HP PRODESK 400 G6 MINIHOMELAB-APPSDocker Compose · this site and other apps

LAN connection SITE-TO-SITE VPN Service / node

Simplified diagram. I've left out IP addresses and the details of the remote network.

DNS · DHCP · MONITORING

homelab-red

ThinkCentre M715q Tiny · Debian 13

AdGuard Home for DNS and DHCP; Prometheus 3.8.0, Grafana 12.3.0, Node Exporter 1.10.2, Blackbox Exporter 0.27.0 and Uptime Kuma for monitoring.

CONTAINERS

homelab-apps

HP ProDesk 400 G6 Mini · Debian 13

Docker Compose for the portfolio and personal applications. Application services are separated from the node dedicated to DNS and monitoring.

NETWORK

Network services

Movistar HGU · Xiaomi AX3600 / OpenWrt · Gigabyte switch

Main LAN with DHCP centralized on the homelab and a LAN-to-LAN VPN to a remote network. Addressing and access details are omitted from this public version.

PROJECTS

Four projects, in detail.

These are real projects I worked on. I don't name customers or share internal data; everything else is told as it happened.

DESIGN & BUILD · VMWARE 8

A new datacenter from scratch

We needed a new datacenter to move customers into from the old one, which ran vSphere 7. I designed and built it: five Dell hosts with 64-core AMD EPYC CPUs, Dell switches at 100 Gb and a Pure Storage array over NVMe over RoCE. A single cluster with its own vCenter.

WHAT I DID

All of the VMware side: compatibility matrix, remote-management firmware, BIOS, drivers, ESXi 8 image, vCenter and cluster. I also initialized the array, working with Pure, and presented the SAN to the hosts. Each host has 200 Gb for SAN and another 200 Gb for vMotion, the array has 400 Gb, and Internet egress is 10 Gb through Fortinet. Management and server consoles sit on their own VLAN, reachable only over VPN. Every customer has its own VDS, and the larger ones also get their own virtual Fortinet.

MIGRATION & RESULT

I migrated whole customers, mostly with a full Veeam backup restored into a new VLAN, keeping their segmentation. I decided which customers moved and when. Some later moved on to Cloud Director. The new CPUs and memory improved performance, and backups and file restores got faster. Backups go to a second Pure array, and I added a Veeam proxy with a per-host affinity rule to spread backup and restore load.

ESXi 8vCenterDellPure StorageVeeamVDS

MIGRATION · SAN → vSAN

SAN to vSAN without changing the network

A seven-host cluster on a Dell storage array with mixed versions: four hosts on vSphere 6.5 and three on 7.0. It had to move to vSAN on local disks without touching customer networks.

HOW WE DID IT

In phases. We added local disks to the hosts and started with three: we upgraded them, built vSAN on them and began moving VMs. The source cluster wasn't in vCenter, so live migrations were done with replication. Every host that freed up was upgraded and joined vSAN, until the new 7.0.3 cluster was complete. Same networks and same VLANs throughout.

THE TRICKY PART

The Distributed Switches had been created on 6.5 and had to be upgraded to 7.0.3 along the way. Beyond that, the only downtime was short, when each replication was finalized.

vSphereSANvSANReplicationVDS

MIGRATIONS · MULTI-DATACENTER

Moving customers between platforms

Not everything was building from scratch. A big part of the job was moving things: whole customers between vCenters, data between two arrays and VMs from Azure to VMware.

WHAT I DID

Depending on the case I used Cross vCenter vMotion, replication or a full Veeam backup restored at the destination. I also built DR between two separate datacenters, with HA for the VMs and replication rules.

SCALE

In total, more than 100 customers onboarded, run or migrated, and around 1,000 virtual machines running 24/7.

Cross vCenterVeeamAzureVMware

DISASTER RECOVERY · VEEAM 13

Veeam DR, actually tested

Disaster recovery between a customer environment and ours: ten VMs, one of them about 3 TB, with RPO and RTO set for each one.

WHAT I DID

I deployed the Veeam 13 components and proxies, the certificates, and configured the backup jobs, the destination and the recovery VLAN. The customer's environment ran entirely on IP addresses, so I created a DNS zone with its records and moved it to names. In our vCenter the customer had a user that could only see its own folder: no hosts, no networks, nothing else.

THE TEST

We brought the VMs up on our side and changed the local firewall rules to point to our infrastructure. The tests went well.

Veeam CDPProxiesDNSPermissionsDR

See the LinkedIn post ↗

NOTES