Director of Systems & Engineering · SRE / DevOps · AI in production · Valencia, Spain
I lead the teams that keep production platforms running: Kubernetes, infrastructure as code, observability, security, and AI that stays inside the client's walls. I come from SRE, so I care less about the demo and more about the system still working at 3 a.m.
- Plan and run DevOps/SRE and development teams (25+ engineers) for internal and client projects
- Led ISO 27001 and ENS Medio certification (2025), now leading the move to ENS Alto (2026)
- Design architectures end to end, including air-gapped servers for hospitals so data never leaves the building
- Built an LLM proxy that switches between local models and commercial APIs without sensitive data leaving the client
- Author of Rune-stone, an open-source spec pattern for consistent AI-generated code
- Platform: Kubernetes, Helm, Argo CD, Terraform, Ansible, GitLab CI, GitHub Actions
- Observability & security: Prometheus, Grafana, Loki, OpenTelemetry, Vault
- Data: PostgreSQL, Kafka, MongoDB, TimescaleDB, FIWARE
- Cloud: AWS, GCP, Azure, Hetzner
- AI: vLLM, Ollama, RAG, MCP, Claude Code
- Languages: Python, Bash, Go, TypeScript
- Tips and good practices for Dockerfiles (part 1)
- Tips and good practices for Dockerfiles (part 2)
- Infra monitoring with Prometheus, Node Exporter and Grafana
- Spanish versions and more on the Pod SRE blog
- Website: podsre.com
- LinkedIn: victorpilestemporal
- Medium: @victorpiles