Open to new opportunities

Mehadi Hasan Shuvo

>DevOps EngineerBackend EngineerSystems & Infrastructure

I build resilient, automated infrastructure — and the backends that run on it.

View my work About me Resume
Uttara, Dhaka, Bangladesh
5+
Years of experience
5
Languages I code in
15+
Production systems shipped
4
Cloud platforms
01

About

DevOps & Backend Engineer with 5+ years working where infrastructure meets software — I build and operate the systems that keep applications reliable, automated, and observable.

I started in backend — Python, Flask, FastAPI and distributed queues — shipping services like a high-availability SMS middleware, then moved to the infrastructure that runs them: containers, Kubernetes, GitOps and CI/CD.

Now at NexGen Cloud, I own production deployments and platform automation while pursuing an M.Sc in CSE at BUET.

  • Own and automate production deployments end-to-end with Ansible & GitOps
  • Build backend services — Python, FastAPI, Flask, Celery, distributed queues
  • Run containers & Kubernetes with ArgoCD, Helm and CI/CD pipelines
  • Stand up observability with Grafana, Prometheus, Loki, Sentry & Alloy
Mehadi Hasan Shuvo
● DevOps Engineer @ NexGen Cloud
02

Experience

From backend engineering into the systems that run it.

DevOps Engineer · NexGen Cloud

Aug 2024 — Present
  • Provide DevOps support and own weekly production releases across Hyperstack, AI Studio, InfraInsight & the Investor Portal — smooth, error-free rollouts.
  • Automated production deployments with Ansible across Infrahub APIs, core services, AuthUI, OpenStack workers, DB backups and smoke tests, triggered by Gitea one-click workflows.
  • Own AI Studio end-to-end across dev, staging, preprod & production with environment-specific CI/CD pipelines.
  • Led the AuthUI + Authentik authentication platform — OAuth (Microsoft, GitHub), 2FA, whitelabel logins and SSO across Gitea & Azure AD.
  • Automated build & publish for five Hyperstack SDKs (Python, Java, Go, JS, TS) with single-tag releases to PyPI and npm.
  • Built observability with Grafana, Prometheus, Loki, Alloy & Sentry, and added SonarQube scanning to CI across environments.

Sr. Software Engineer · Intercloud Limited

Apr 2024 — Jul 2024
  • Delivered Managed Kubernetes-as-a-Service — one-click cluster deployment through the Brilliant Cloud portal with stateful-app support on OpenStack Cinder.
  • Implemented a GitOps + ArgoCD CI/CD solution for Brilliant Cloud.
  • Deployed and maintained full-stack applications on Docker & Kubernetes, exposed to the internet over SSL.

Software Engineer · Intercloud Limited

Oct 2022 — Mar 2023
  • Built a high-availability SMS middleware (Flask, Celery, MongoDB) on Kubernetes, plus an SMS-to-voice conversion service.
  • Developed the Brilliant Drive backend and Brilliant EDMS on S3-compatible object storage, adding a Reminder module.
  • Added WebSocket support to the Brilliant load balancer for efficient real-time traffic.
  • Refactored the Brilliant Cloud Invoice Service for automated monthly billing, and set up Loki + Grafana monitoring.

Trainee Software Engineer · BJIT Group

Jan 2022 — Oct 2022
  • Started my software career here — learned the Agile software development process and delivery workflow.
  • Built a small Android application in Java as a hands-on training project.
03

Tech Stack

The tools I reach for across infrastructure, backend and observability.

DevOps & Infrastructure

DockerKubernetesAnsibleTerraformArgoCDHelmGitOpsJenkinsGitHub ActionsGitLab CI

Observability

GrafanaPrometheusLokiSentryElasticsearch

Languages

PythonGoJavaScriptTypeScriptJavaC++

Backend

FastAPIFlaskDjangoSpring BootExpressReact

Data & Messaging

PostgreSQLMySQL / GaleraMongoDBRedisRabbitMQKafkaCelery

Cloud

OpenStackAWSGCPDigitalOcean
04

Selected Works

Everything I've designed, built and shipped.

NexGen Cloud

11 projects

Production Deployment Ownership

I own every production release at NexGen

4 products

Context · Production releases across four core products needed a single, reliable owner and full automation.

What I did · Own and execute all production deployments across Hyperstack, AI Studio, InfraInsight and the Investor Portal — Hyperstack through one-click PR-merge Ansible/Gitea workflows, the others via Docker-based Gitea workflows. Built the full Ansible deployment automation behind it: Infrahub APIs, core services, the Admin Portal, AuthUI, database backups, OpenStack workers, API docs, proxy services and smoke/integration tests.

Centralized, low-friction and auditable production releases across four products — reliable, repeatable and fully automated.

AnsibleGiteaCI/CDRelease ManagementSmoke Tests

AuthUI + Authentik Platform

Modern authentication, end-to-end

SSO · 2FA

Context · Hyperstack needed a modern, extensible authentication system for internal staff, external customers and whitelabel partners.

What I did · Led the end-to-end implementation of AuthUI — from external-contractor coordination to CI/CD and production rollout. Integrated Microsoft & GitHub OAuth, added 2FA, and shipped a fully whitelabeled solution letting partners run custom-branded logins on the Hyperstack backend. Ran phased Authentik upgrades and extended SSO across Gitea, Azure AD and internal AI tooling (OIDC for Open WebUI / Ollama).

A scalable, secure, partner-friendly auth platform used across internal staff, customers and partner companies.

AuthentikOAuth2FASSOOIDCWhitelabelCI/CD

Observability Stack

Metrics, logs & tracing across every environment

all envs

Context · Debugging and monitoring were inconsistent across VMs, Kubernetes and multiple OpenStack regions.

What I did · Built a containerized Grafana stack (Grafana, Prometheus, Loki, Alloy), exposed via Caddy with OAuth, and migrated agents from Promtail/Node Exporter to Grafana Alloy with in-log sensitive-data redaction. Built Caddy reverse-proxy dashboards and provisioned alert rules via the Grafana API. Designed a streaming backup of Loki & Prometheus data to Ceph RadosGW with zero local-disk overhead. Deployed self-hosted Sentry for APM & error tracking.

Standardized observability and faster debugging across dev, staging, preprod and production.

GrafanaPrometheusLokiAlloySentryCephCaddy

Hyperstack SDK Automation

Single tag → published SDKs in 5 languages

5 languages

Context · Publishing SDKs across five languages was manual, slow and error-prone.

What I did · Automated build & publish workflows for Python, Java, Go, JavaScript and TypeScript SDKs — single-tag-driven releases to PyPI and npm, with GitHub Actions copying generated SDKs from the generator repo to each language repo after verification.

Eliminated manual release effort, improved consistency and accelerated developer adoption.

GitHub ActionsPyPInpmCI/CDAutomation

AI Studio — End-to-End Ownership

Owned a product from dev to production

dev → prod

Context · AI Studio needed a single owner across every environment and a fully automated delivery path.

What I did · Took full ownership across dev, staging, preprod and production. Designed environment-specific CI/CD pipelines, set up separate Pritunl VPN access for secure connectivity, merged the AI Studio URL into Hyperstack, and built CI/CD for inference servers, Watchtower and integration tests — plus user guides and service docs.

Transformed AI Studio into a fully automated, reliable, end-to-end platform with minimal manual intervention.

CI/CDDockerWatchtowerPritunl VPNIntegration Tests

Cost & Resource Automation

Real-time orphan-resource recovery

Instant visibility into revenue-impacting waste, healthier disks and direct cost optimization.

KubernetesPythonHelmSlack

Release Pipeline Automation

Self-service, self-healing deployments

Higher deployment velocity, fewer bottlenecks and better recovery from failed releases.

CI/CDAnsibleGiteaSlack

Ephemeral Environments

On-demand OpenStack worker pods

Lower idle cost, better elasticity and faster turnaround for on-demand workloads.

KubernetesOpenStackAutoscaling

Code Quality & Security Automation

SonarQube scanning in CI

Improved code quality, security posture and developer accountability.

SonarQubeCI/CDSecurity

OpenStack Worker Automation

GitOps automation & logging for OpenStack workers

Consistent, GitOps-driven worker deployments with centralized logging across regions.

ArgoCDGitOpsHelmElasticsearch

Production Incident Response

Root-cause analysis under pressure

Faster resolution, reduced downtime and higher confidence during deployments.

ElasticsearchKibanaIncident Responsegunicorn

Intercloud Limited

6 projects

Managed Kubernetes-as-a-Service

One-click clusters on the Brilliant Cloud portal

1-click clusters

Context · Brilliant Cloud users needed self-service Kubernetes without touching the underlying infrastructure.

What I did · Built a Managed Kubernetes service with an Ansible-driven provisioning flow — one-click cluster deployment from the portal, with stateful-app support backed by OpenStack Cinder — and implemented a GitOps + ArgoCD CI/CD solution for Brilliant Cloud alongside it.

Self-service cluster deployment for Brilliant Cloud customers.

KubernetesAnsibleOpenStackArgoCDGitOps

SMS Middleware (HA)

High-availability A2P messaging service

HA on K8s

Context · Brilliant SMS needed to connect its portal to an A2P SMS aggregator with production-grade reliability and throughput.

What I did · Built the SMS middleware with Flask, Celery and MongoDB — Celery workers giving asynchronous, retry-safe delivery — deployed as a high-availability production environment on Kubernetes, plus a middleware service for SMS-to-voice conversion.

Reliable, scalable SMS delivery running in production for the Brilliant SMS platform.

FlaskCeleryMongoDBKubernetesA2P

Brilliant Drive Backend

Object-storage-backed file service

Shipped a new storage service within the Brilliant Connect ecosystem.

PythonS3Object Storage

LoadBalancer WebSocket Support

Real-time traffic for the Brilliant LoadBalancer

Efficient real-time traffic handling across balanced backends.

WebSocketLoad BalancingNetworking

Brilliant EDMS

Document management + reminders

Expanded EDMS with storage and scheduling capabilities.

FlaskObject StoragePython

Invoice Service Automation

Automated monthly billing

Hands-off monthly billing and clearer backend observability.

PythonBillingLokiGrafana
05

Education & Recognition

Education

M.Sc in CSE (Part-time)
Bangladesh University of Engineering & Technology (BUET)
Ongoing
2023 — Present
B.Sc in CSE
Chittagong University of Engineering & Technology (CUET)
CGPA 3.55 / 4.00
2017 — 2022
Higher Secondary Certificate (HSC)
Rajuk Uttara Model College
GPA 5.00 / 5.00
2016
Secondary School Certificate (SSC)
Rajuk Uttara Model College
GPA 5.00 / 5.00
2014

Publications & Talks

Publication

“A Smart Prediction Model for Identifying Homeless People Using Machine Learning…” — Int'l Journal of Society Systems Science (Inderscience, 2023)

View
Conference Session

Speaker — “Cloud in Digital Bangladesh”, UIU Computer Club (Aug 2023)

// what colleagues say

“Dependable and proactive — Mehadi takes real ownership of our production systems, automates the hard parts, and keeps deployments reliable.”
SMShahed MehbubHead of DevOps, NexGen Cloud
“Efficient, effective, and an absolute pleasure to work with. I wish Shuvo all the best.”
SISimon IslamSolutions Architect, NexGen Cloud
“Working with Shuvo has been a pleasure. His coding expertise and problem-solving skills are truly commendable.”
JMJahidul MominBackend Developer, NexGen Cloud
“His software engineering skills are truly impressive — thorough, reliable and detail-oriented.”
MNMushfikunnabi NijhumSoftware Engineer, Intercloud Limited
06

Get in touch

Have a role, a project, or an infrastructure problem worth solving? Let's talk.

Email[email protected]Phone+8801777267000
LocationUttara, Dhaka, Bangladesh