Skip to content
View iacker's full-sized avatar
⚔️
⚔️
24 followers · 54 following

Achievements

Achievement: Pull Sharkx2Achievement: Pair ExtraordinaireAchievement: YOLOAchievement: QuickdrawAchievement: Starstruck

Achievements

Achievement: Pull Sharkx2Achievement: Pair ExtraordinaireAchievement: YOLOAchievement: QuickdrawAchievement: Starstruck

Block or report iacker

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.

Content in all repositories owned by your account will be closed.
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
iacker/README.md
iacker

Cloud & Platform Engineer, DevSecOps

GPU serving · Kubernetes · hybrid IaC · DevSecOps · FinOps



Kepler maintainer (CNCF) · measuring energy consumption in Kubernetes


views followers

What I do

I build and operate the infrastructure that serves GPU and LLM workloads in production, on hybrid cloud (on-prem plus AWS/Azure). I automate provisioning, secure the chain, and keep cost under control.

  • Hybrid cloud : AWS, Azure, on-prem, GPU bare metal and cloud
  • Kubernetes / Platform : k3s, Talos, Helm, GitOps (Flux, Argo), autoscaling
  • IaC : Terraform, Ansible, reproducible pipelines, clean teardown
  • DevSecOps : eBPF/NDR, hardening, supply chain, zero trust
  • GPU / LLM serving : vLLM, FP16/FP8 quantization, DCGM, measured €/token cost

Skills


Projects

Project In short
gpu-inference-reliability-lab SRE lab: vLLM on k3s, DCGM observability, measured €/token and energy, incident runbooks.
Talos_Bastion_DevSecOps Talos Linux GitOps cluster: Cilium (eBPF), Traefik, ArgoCD, zero-trust access.
terraform-runpod-vllm One terraform apply, a RunPod GPU serves an LLM over the OpenAI API. Budget guard, clean teardown.
mcp-scalpel Semantic filtering proxy for the Docker MCP Gateway. Cuts catalog tokens per turn.
Azure-Pipeline-Custom-Image Azure DevOps CI/CD: custom VM image deployed via Managed DevOps Pool.

Contact

LinkedIn Email

Pinned Loading

  1. gpu-inference-reliability-lab gpu-inference-reliability-lab Public

    Lab SRE pour l'inférence LLM sur GPU : k3s, vLLM, Kueue, observabilité DCGM, coût €/token et énergie mesurés. Runbooks d'incidents vécus.

    Python 1

  2. Talos_Bastion_DevSecOps Talos_Bastion_DevSecOps Public

    My talos linux bastion, gitops, ndr ebpf, security scanning

    2

  3. Azure-Pipeline-Custom-Image Azure-Pipeline-Custom-Image Public

    Pipeline to install and deploy a custom VM image with the Managed Devops Pool

    1

  4. terraform-runpod-vllm terraform-runpod-vllm Public

    Déploie un LLM sur GPU RunPod en vLLM (API OpenAI) en un terraform apply — protection anti-doublon false-500 et budget guard

    Python 1

  5. mcp-scalpel mcp-scalpel Public

    Proxy local de filtrage sémantique pour le Docker MCP Gateway : réduit les tokens de catalogue MCP par tour (35-52% selon le hint). Pas de cloud.

    Python 1

  6. Harness-Cluster Harness-Cluster Public

    Kubernetes GitOps cluster with specialised harnesses

    1