Skip to content
View pablodelarco's full-sized avatar
🎯
Focusing
🎯
Focusing
9 followers · 21 following

Achievements

Achievement: YOLOAchievement: StarstruckAchievement: Pull Sharkx2Achievement: Pair ExtraordinaireAchievement: Quickdraw

Achievements

Achievement: YOLOAchievement: StarstruckAchievement: Pull Sharkx2Achievement: Pair ExtraordinaireAchievement: Quickdraw

Block or report pablodelarco

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
pablodelarco/README.md

👋 Hi, I'm Pablo

Pablo del Arco is an AI Cloud Engineer: he builds AI agents, MCP servers and RAG systems, plus the Kubernetes infrastructure they run on.

He works as a Technical Project Manager at OpenNebula Systems, where he shipped the EuroCopilot AI stack across 5 European cloud zones and now leads the EU research projects EMERALD and CASANDRA.

On the side, he automates real businesses with AI agents: a 9-property rental business runs unattended on a Raspberry Pi. He also writes about Kubernetes and AI on Medium, where his homelab series was featured by the Cilium community.

Learn more on pablodelarco.com →

README · A bit more about me

About me

  • Based in Valencia, Spain. Before that: Helsinki as a 5G engineer at Nokia, and a double MSc at EURECOM (France) and Aalto University (Finland).
  • Everything I ship gets tested on my two-node Kubernetes homelab first: 25+ self-hosted apps run entirely through GitOps.
  • Certified: OpenNebula Expert, plus Claude Code in Action and Model Context Protocol: Advanced Topics, both by Anthropic.
  • Off keyboard: acoustic and Spanish guitar, padel and football.

How I work

  • Every project starts in Claude Code, and it starts with a spec, not with code: a structured interview skill turns the idea into a written design doc first.
  • The same skill set enforces best practices from day one in every repo: plan before build, independent reviewer agents that grade each change, and a verifier that has to pass before anything ships.
  • Nothing reaches production untested: it runs on my own Kubernetes cluster first, documented and monitored.

Featured projects

  • kubernetes-homelab: GitOps homelab on K3s with ArgoCD, Cilium and Longhorn.
  • finetwork-mcp: MCP server for a Spanish telecom. Query invoices, services and billing from Claude Code or any MCP client.
  • opennebula-helm: production-ready Helm chart for OpenNebula.

Writing and talks

Contact

Pinned Loading

  1. pablodelarco pablodelarco Public

    Config files for my GitHub profile.

  2. kubernetes-homelab kubernetes-homelab Public

    Resources and configurations for building a Kubernetes homelab powered by K3s with GitOps workflows.

    Shell 31 5

  3. flower-opennebula flower-opennebula Public

    Flower federated learning integration for OpenNebula marketplace

    Shell

  4. finetwork-mcp finetwork-mcp Public

    MCP server for Finetwork (Spanish telecom) — query invoices, services, billing from Claude Code or any MCP client

    TypeScript 1

  5. opennebula-helm opennebula-helm Public

    Production-ready OpenNebula Helm Chart for Kubernetes with auto-updating Docker image

    Shell 1

  6. oneapps-nemoclaw oneapps-nemoclaw Public

    NemoClaw OpenNebula Marketplace Appliance - NVIDIA AI agent security stack with GPU passthrough

    Shell