| Currently | DevOps Intern at Springer Capital · Remote (Chicago, USA) |
| Foundation | Git & GitHub workflows · Linux & Bash · Docker · CI/CD |
| Focus | Automating reliability: pipelines, health checks, logging and alerting |
| Cloud | AWS: EC2 · ECS Fargate · ECR · VPC · IAM · Load Balancers · Security Groups |
| Now Learning | Kubernetes · Terraform · HashiCorp Nomad · DevSecOps |
| Philosophy | Reliability isn't an accident. Automate it, observe it, then keep improving it. |
| Open to | DevOps Engineer · Cloud Engineer · SRE · Platform Engineer |
| Based in | Dhaka, Bangladesh · working with engineers across countries |
Every project in this profile started the same way: something broke, and I wrote down why.
| Chapter | What happened |
|---|---|
| 01 · The Question | Tutorials end at "it works on my machine." I wanted to know what comes after: how code becomes a live, reachable, reliable service. So I started with the craft: Git, pull requests, merge conflicts, SSH and Linux. |
| 02 · The First Pipeline | Mini Todo API. A FastAPI app, a pull-request workflow and a GitHub Actions pipeline running tests, dependency checks and security checks. I debugged every failed run until it went green, then deployed to EC2 as a systemd service. The app was the practice. The pipeline was the point. |
| 03 · A Real Server | I took a mentor's MERN incident-management app and got it running on AWS EC2, through SSH timeouts, a pending kernel upgrade, processes dying after disconnect and Git auth failures. I learned to debug one layer at a time with nc, curl and logs, instead of changing settings at random. |
| 04 · Containers to the Cloud | Cloud Health Monitor. Dockerized, gated by GitHub Actions, shipped through ECR to ECS Fargate, with an email alert on every run. Then production pushed back, three times (see the incident log). |
| 05 · The Team | Aug 2026: DevOps Intern at Springer Capital, working remotely with engineers in different countries. Built an event-driven quiz pipeline with n8n, evaluated n8n in an isolated staging environment and audited Grafana Loki for high-cardinality labels (Jira DEV-290). |
| 06 · What's Next | Making reliability automatic at a bigger scale: Kubernetes, Terraform and DevSecOps, earned through real projects instead of copied stacks. |
🗓️ Milestones
| When | What |
|---|---|
| Aug 2026 – Present | DevOps Intern @ Springer Capital · Remote (Chicago, USA): Linux, Docker, CI/CD, AWS, Grafana Loki and n8n automation |
| 2026 | DevOps and Cloud Engineering certification track @ bongoDev |
| Nov 2025 – Feb 2026 | Junior Executive @ SkyTech Solutions: client communication and lead handling |
| Dec 2021 – Dec 2025 | B.Sc. in Computer Engineering @ BUBT |
flowchart LR
A(["💻 git push"]) --> B{{"⚙️ GitHub Actions<br/>lint · test · scan · audit"}}
B -- "❌ any failure" --> X(["⛔ Blocked + 📧 email alert"])
B -- "✅ all green" --> C["🐳 Docker build"]
C --> D[("📦 Amazon ECR")]
D --> E["☁️ ECS Fargate"]
E --> F(["🩺 Live health dashboard"])
# production-mindset.sh
$ ship every change # push → lint → test → build → deploy
$ keep it reliable # systemd · safe rollouts · stable endpoints
$ keep it visible # Loki logs · uptime checks · alerts
$ debug before guessing # nc · curl · Security Groups · logs|
Docker → GitHub Actions → ECR → ECS Fargate A real-time dashboard for live CPU, memory and disk usage, plus a URL uptime checker. Every push runs ESLint, Prettier, Jest (with a coverage threshold), a security scan,
|
A MERN app on AWS EC2 The app is my mentor's. The deployment, infrastructure and ops are mine: Ubuntu 24.04, Security Groups, SSH key pairs, EC2 Instance Connect and MongoDB under systemd. Along the way I debugged SSH timeouts, a pending kernel upgrade, dying processes and Git auth failures.
|
|
Pull request → tests → security checks → EC2 My first full Git-to-deployment workflow. A FastAPI app with a PR-based flow, GitHub Actions for tests, dependency checks and security checks, and secrets kept out of the repository. It deploys to EC2 as a systemd service, and the live API is verified through its endpoints.
|
Baserow → Webhook → n8n → Mattermost An event-driven workflow built at Springer Capital. A quiz submission in Baserow triggers a webhook, n8n maps and transforms the fields, and a bot posts a structured result to Mattermost. It comes with a scoring script, answer-key notes and workflow screenshots.
|
🧪 More from the lab
- 🔧 DevOps Deployment Pipeline: a full workflow with Linux, Git, Docker, CI/CD, HashiCorp Nomad and Grafana Loki for monitoring.
- ⚙️ GitHub Actions Testing: automated testing pipelines for JavaScript projects.
- 🎮 Antigravity Tic-Tac-Toe: a Node.js terminal game with an AI opponent, Jest tests and a full GitHub Actions CI pipeline.
- 🔀 Git & GitHub Workflow: branching, merging, pull requests, conflict resolution and CI/CD exercises.
- 🔗 QR Code Generator: a small QR code API built with Go.
A green pipeline means nothing until the app is reachable in production. These are the three times I learned that the hard way, all from Cloud Health Monitor.
| 🔥 What broke | 🔎 Root cause | 🛠️ Fix | 💡 Lesson |
|---|---|---|---|
| 1,000+ CI errors in one run | An AWS CLI folder was accidentally committed, so the style checks flagged files that were never part of my code | Removed it from Git and added it to .gitignore |
Keep noise out of the pipeline, or real failures get buried |
| App unreachable after a green deploy | A Security Group rule was blocking public traffic | Opened the correct inbound port | Green CI is not the same as reachable. Verify the network path |
| "Dead" app after every restart | Fargate assigns a new public IP on each restart, so I was testing an IP that no longer existed | Switched to the stable ECS service URL | Never depend on an address that changes |
| 🐳 Containerization Docker workflows |
📈 Observability Grafana Loki logging |
⚙️ Orchestration HashiCorp Nomad |
☁️ Cloud AWS EC2 deployments |
🌱 Leveling up:
| ☁️ Cloud & Containers | AWS (EC2, ECS Fargate, ECR, VPC, IAM, Load Balancers, Security Groups) · Docker |
| 🔄 CI/CD & Version Control | GitHub Actions (lint, test, security and deploy gates, email alerts) · Git · GitHub · pull requests · SSH auth |
| 📈 Observability | Grafana Loki (stream and label-cardinality audits) · cron monitoring · uptime and health checks · Mattermost notifications |
| 🔐 DevSecOps | GitHub secrets · secret scanning · CodeQL · npm audit · eslint-plugin-security · SSH hardening (ED25519, key-only access) |
| 🤖 Automation | Ubuntu · systemd · Bash · n8n workflows · webhooks · REST APIs · Python (FastAPI) · Node.js · Go |
|
CI/CD pipelines with quality and security gates, Docker images and automated rollouts. |
AWS deployments on EC2, ECS Fargate and ECR, with VPC, IAM and Security Group awareness. |
Health checks, logging, alerting and a written record of every failure and fix. |
Automation workflows and repeatable pipelines: n8n, webhooks, REST APIs and scripts. |
Contributing to bongoDev/git-and-github, a community learning resource for Git & GitHub.
I'm looking for DevOps, Cloud, SRE and Platform Engineering opportunities where I can ship, operate and improve production systems, and automate the parts that keep them up at 3 a.m.
🛠️ Build it. Ship it. Watch it. Improve it.