Staff SRE · Kubernetes · AWS
I build cloud-native systems that don't break — from bare metal to the cloud.
nabeem.com · LinkedIn · Medium · Email
nabeem@sre:~$ whoami
Staff Site Reliability Engineer — 5+ years keeping production up, from bare-metal
datacenters to Kubernetes and AWS. I care about reliability, platform automation,
GitOps, observability, incident response, and infrastructure that's cheap to run.
nabeem@sre:~$ cat now.md
Staff SRE @ Zscaler — building and scaling the bare-metal fleet behind a global
cloud-security platform.
nabeem@sre:~$ cat learning.md
Diving into AI infrastructure — GPU scheduling, and the architecture behind training
and serving models at scale.
nabeem@sre:~$ cat fun-fact.txt
An athlete and fitness enthusiast at heart; a computer science engineer by profession.| Project | What it is |
|---|---|
| Bare Metal K8s · case study ↗ | Netboot blank machines into a k3s cluster in one command — under UEFI Secure Boot with keys I mint myself. My own PK/KEK/db sign the whole shim → GRUB → kernel chain: DHCP → UEFI HTTP Boot → autoinstall → Ansible → k3s. |
| EKS GitOps Platform · case study ↗ | A production-style Amazon EKS platform — Terraform + ArgoCD App-of-Apps + Karpenter + full observability, backup, and Kyverno policy across ~30 components, all reconciled from one repo. |
| Namespace Bootstrap Operator · case study ↗ | A Kubernetes operator that turns one NamespaceProfile resource into a governed namespace — ResourceQuota, LimitRange, NetworkPolicies and Pod Security labels — then keeps it that way, correcting drift on a watch event or refusing it outright at admission. Go, Kubebuilder, CEL, server-side apply. |
| Reach · case study ↗ | Give AI agents policy-controlled access to every machine you own — without SSH, VPNs, or open ports. Go, MCP, zero-trust, approval workflows. |
The tooling I reach for to build, secure, and operate systems at scale.
nabeem@sre:~$ from bare metal to the cloud.
