I’m a Site Reliability Engineer with 10 years of experience operating infrastructure that can’t quietly fail, from on-prem virtualization and storage to high-volume observability platforms and Kubernetes.

I work at the intersection of reliability engineering, observability, automation, and AI-assisted operations. My background in systems, networking, and on-prem infrastructure gives me a broader operational view than cloud-only DevOps profiles; my current focus is on observability pipelines, SLO-driven engineering, and practical AI tooling for operations teams.

What I do

  • Observability at scale: Grafana, Loki, Thanos, Vector, CloudWatch - stack design, query performance, SLO frameworks, platform hygiene.
  • Platform engineering: Kubernetes operations, Helm, Argo CD, CI/CD (Jenkins, GitLab), Terraform and Ansible for repeatable delivery.
  • On-prem & hybrid infrastructure: Proxmox, Ceph, multi-datacenter HA, migrations from legacy virtualization.
  • AI-assisted SRE tooling: Graphia, a domain-specific SRE agent for Grafana diagnosis - RBAC-aware, MCP-based, built for real operations rather than demos.

Open source

I build tools around real operational pain points and keep them open. Selected projects:

  • SSHplex - Modern SSH multiplexer with multi-source inventory (NetBox, Ansible, Consul) and tmux/iTerm2 backends. Built to replace expensive legacy tooling.
  • terraform-provider-centreon - Go-based Terraform provider for Centreon API V2. Turns monitoring configuration into infrastructure as code.
  • OpenClaw Audit TUI - Terminal UI for auditing agent sessions, tool usage, and live events with streaming to Discord/Telegram/Slack.

More on the Open Source page.

Background

I’ve been operating infrastructure since I was 15, starting with a multi-node Minecraft platform on OVH dedicated servers. Since then: a Master’s degree project on BGP + ElasticSearch + ML + JA3 signature analysis for DDoS auto-mitigation, then 10 years across system administration, virtualization, on-prem storage, and SRE work on production platforms ingesting multiple TB of observability data per day.

Contact

Email: [email protected] · Paris area / remote · Available for freelance missions.