Islamabad, Pakistan · Remote

The Tech Tatler

Platform · Security · Reliability

Muhammad Shahab Alam

Muhammad Shahab Alam

Quick Facts

DevOps Engineer
Islamabad, Pakistan · Remote
BS Cybersecurity, FAST
ISC2 Certified in Cybersecurity

I build cloud infrastructure that stays secure and boring in production.

DevOps engineer working across Google Cloud and AWS, with a background in cybersecurity.

By Shahab Alam

I studied cybersecurity before I moved into DevOps, and it still shapes how I work. Access should be given on purpose. Releases should be traceable. Anything that doesn't need to be public shouldn't be.

Day to day, that means a lot of Terraform, Kubernetes, and the less visible parts of running production: permissions, networking, alerts, and docs that people actually read.

Currently At

Jul 2026 – Present

DevOps Engineer

Sapience AI

Sapience AI is the collective intelligence platform for professional communities. I work on its cloud infrastructure, access control, and the shared services behind its AI features.

A few results from production work

35%
lower cloud costs
70%
faster deployments
50%
faster incident detection
2M+
records migrated with ~5 min downtime
01

Experience

What I've worked on, and what got better because of it.

Jul 2026 – Present

Remote

Current

DevOps Engineer at Sapience AI

Sapience AI is the collective intelligence platform for professional communities.

I look after our Google Cloud and Workspace setup: who can access what, how our networks fit together, the environments customers run in, and the shared services our AI features depend on. I also review infrastructure changes and help newer engineers find their footing.

  • Set up GPU-backed model serving on Kubernetes that other services reach only over private networking. All of it is defined in Terraform.
  • Helped launch dedicated customer environments, with a release pipeline that only deploys signed, verified images.
  • Moved cloud access to group-based roles with time-limited elevated access, and brought most hand-built resources under Terraform.
  • Wrote an offboarding tool that removes someone's access across our cloud and SaaS tools, then checks that it's really gone.
  • Rebuilt our self-hosted CI runners so every build starts clean and carries no long-lived credentials. Queue times dropped about 30%.
  • Added LLM tracing with Langfuse, tidied up alerting, and wrote the runbooks and onboarding docs the team now uses.

Oct 2025 – Jul 2026

Islamabad

DevOps Engineer at Centrox AI

  • Cut AWS costs about 35% by moving container workloads to Fargate, with Spot capacity for staging.
  • Made deployments about 70% faster by switching from full service recreation to rolling updates.
  • Split staging and production into separate Terraform environments and automated DNS.
  • Set up private connectivity to a managed database using AWS PrivateLink.
  • Helped move 2M+ records across regions with about five minutes of downtime and no data loss.
  • Built Prometheus and Grafana monitoring that caught incidents about 50% sooner.
  • Ran self-hosted GitLab runners and code-quality checks, and taught developers to handle routine deploys on their own.

Jul – Sep 2024

Islamabad

DevOps Intern at iEngineering

  • Monitored security events with a Wazuh SIEM and helped shorten incident response.
  • Set up EC2 auto scaling, load balancing, and Lambda-based alerting on AWS.
02

Projects

A mix of things I built at work and on my own.

Work

GPU model serving on Kubernetes

A shared service for running embedding models on GPUs. It isn't exposed to the internet, other services reach it privately, and the whole setup lives in Terraform.

Private by default, reproducible from code

KubernetesGPUsTerraformPrivate networking
Work

Offboarding automation

Give it one email address and it removes that person's access from every connected tool, then checks each one to confirm nothing was missed.

One command instead of a checklist

AutomationIAMGoogle WorkspaceGitHub
Personal

Infrastructure and delivery pipeline

A Django and React app in Docker behind Nginx. Terraform provisions the AWS network and servers, and GitLab CI promotes builds from staging to production.

About 90% less manual setup

TerraformAWSDockerGitLab CI
Personal

Monitoring and status page

Prometheus and Grafana across several on-prem environments, with alerts for the team and a public status page for people who just want to know if things are up.

Incidents spotted about 50% sooner

PrometheusGrafanaK3s

Side projects

Small products I built and run myself
Kubernetes Visual Reference

Kubernetes Visual Reference

Learn Kubernetes visually.

A one-page visual reference for clusters, nodes, pods, services, deployments, and networking, paired with a three-tier K8s notes app.

Project
Turtle dashboard

Turtle

Track smarter. Apply better.

A minimalist job-application tracker: one clean dashboard instead of messy spreadsheets and scattered notes.

EcoPilot dashboard

EcoPilot

Climate action, simplified.

Helps small and medium-sized businesses measure their carbon footprint and take practical steps to reduce it, with an AI assistant for guidance.

03

Tech stack

The tools I use most, grouped by area.

Google Cloud
AWS
Kubernetes
Terraform
Docker
Helm
GitHub Actions
GitLab
Prometheus
Grafana
MongoDB
Redis
Python
Bash
Linux
Nginx
Google Cloud
AWS
Kubernetes
Terraform
Docker
Helm
GitHub Actions
GitLab
Prometheus
Grafana
MongoDB
Redis
Python
Bash
Linux
Nginx

Google Cloud

GKECloud RunCloud SQLVPCPrivate Service ConnectIAPCloud ArmorLoad BalancingArtifact RegistryBinary AuthorizationPub/SubCloud StorageBigQueryCloud MonitoringIAMVertex AI

AWS

EC2ECSFargateVPCALBAuto ScalingRoute 53IAMLambdaS3CloudWatchPrivateLink

Infrastructure and CI/CD

TerraformGitHub ActionsGitLab CI/CDHelmDocker

Kubernetes

GKEK3sGPU node pools

Security

IAM and RBACcosignTrivySonarQubeWazuhSOC 2GDPR

Observability

PrometheusGrafanaCloud MonitoringLangfuse

Data

OpenSearchMongoDB AtlasRedisApache Airflow

Languages

PythonBashLinuxGitNginx
04

Writing and education

Certifications

  • Certified in Cybersecurity (CC)

    ISC2 · 2024

  • Azure Fundamentals (AZ-900)

    Microsoft

Education

BS Cybersecurity

FAST School of Computing, Islamabad

2021–2025