Hi, I'm Mohammed Zafeeruddin

DevOps & MLOps Engineer | Hyderabad, Telangana

Passionate DevOps and MLOps engineer specializing in infrastructure automation, container orchestration, and ML pipeline deployment. Experienced in building scalable AI platforms, orchestrating Kubernetes clusters, and implementing CI/CD pipelines. I transform complex technical challenges into elegant infrastructure solutions.

Mohammed Zafeeruddin - Software Engineer
Scroll down

About Me

I'm a DevOps and MLOps engineer with expertise in infrastructure automation, microservices architecture, and cloud-native technologies. Currently working as a DevOps/MLOps Engineer, I've architected and built critical infrastructure including Baseer Builder, a low-code/no-code AI platform that's now being used by Eastern Provincial Municipality with 2,000+ cameras and 120+ use cases.

My experience spans from building PoCs for computer vision applications to architecting production-grade Kubernetes-based microservices. I'm passionate about automation, distributed systems, and making complex infrastructure accessible through elegant solutions. From reducing training time from 55 hours to 5 hours through distributed training to automating deployments that saved hours of manual work, I focus on creating impactful solutions.

12+ Microservices
2+ Years Experience
120+ Use Cases

Experience

DevOps & MLOps Engineer

May 2024 - Present
Current Organization
  • Architected backend and microservices infrastructure for Baseer Builder, a low-code/no-code AI platform now used by Eastern Provincial Municipality with 2,000+ cameras and 120+ use cases
  • Currently working with Eastern Provincial Municipality of Saudi Arabia, deploying and maintaining AI-powered solutions across 2,000+ cameras. Previously worked with Dubai and Riyadh airports on critical PoC implementations including automated monitoring systems and real-time camera stream management
  • Developed multiple PoCs including crowd detection models with real-time inference for Male, Female, and Children classification
  • Built people counting systems for mall entrances/exits with live stream inference
  • Created employee availability and unauthorized access detection systems
  • Developed automated cookie refresh system for Dubai Airport PoC using Selenium and SMTP, ensuring continuous camera stream monitoring
  • Initiated work on Baseer Builder platform, laying the foundation for the production system
  • Developed 7 out of 12 persistent microservices in Python, including Model Training Service (MDT), Jupyter Notebook Service (JBN), and real-time monitoring services
  • Implemented distributed training using PyTorch DDP, reducing training time from 55 hours to 5 hours by pooling GPU resources
  • Built automated deployment pipeline using Ansible and Bash scripts with MQTT for real-time metrics, enabling easy machine onboarding with Kubernetes cluster setup
  • Designed CI/CD pipelines using Jenkins and ArgoCD for automated deployment across dev, prod, and EPM environments
  • Architected highly available NGINX infrastructure using Keepalived for failover, ensuring high availability for critical services
  • Dockerized 10+ containers with multi-stage builds and Nuitka optimization, reducing deployment time from 2 hours to 10 minutes
  • Integrated Prometheus, Grafana, and OpenAlerts for comprehensive monitoring and alerting of services and cameras
  • Built HLS streaming service for RTSP stream conversion, optimized for 6 concurrent streams

Skills & Technologies

🐍

Python

Backend development, microservices, automation scripts

☸️

Kubernetes

Cluster orchestration, deployments, service architecture

🐳

Docker

Containerization, multi-stage builds, optimization

πŸ“ˆ

MLFlow

Model tracking, experiment management, model registry

πŸ”§

CI/CD

Jenkins, ArgoCD, GitHub Actions, automation pipelines

πŸ“Š

Monitoring

Prometheus, Grafana, OpenAlerts, observability

πŸ€–

ML/AI

PyTorch, distributed training, model deployment

🌐

Cloud/DevOps

Cloudflare, Ansible, NGINX, infrastructure automation

Featured Projects

Medium Clone

Medium Clone - Full Stack Blogging Platform

A full-stack blogging platform with real-time notifications, interactive comments/replies, and comprehensive blog engagement features. Built with React frontend and HonoJS backend, deployed on Cloudflare Workers and Pages. Features Google OAuth + OTP authentication, PostgreSQL database with Prisma Accelerate, and Redis for caching. Images hosted on Cloudflare R2 with CDN distribution.

React HonoJS Cloudflare Redis PostgreSQL Prisma
k8s-playbooks

k8s-playbooks - Collection of Ansible Playbooks for Kubernetes

Collection of Ansible playbooks to deploy and manage Kubernetes-native applications. Useful for DevOps engineers or SREs looking to automate k8s setups. Includes production-ready playbooks for joining worker nodes, setting up masters, and deploying GitOps and monitoring tooling.

Overview: Production-ready Ansible playbooks for automating common Kubernetes operations. Each playbook is self-contained and documented for easy integration into infrastructure automation workflows.

Ansible Kubernetes DevOps Automation
Baseer Builder

Baseer Builder - Low-Code AI Platform

Production AI platform currently used by Eastern Provincial Municipality with 2,000+ cameras and 120+ use cases. Architected complete backend infrastructure with Kubernetes microservices, including model training, Jupyter notebooks, real-time inference using Triton, and comprehensive monitoring. Features distributed training, automated deployment pipelines, and high-availability infrastructure.

Python Kubernetes Docker PyTorch MLFlow Triton

Get In Touch

I'm always open to discussing new projects, creative ideas, or opportunities to be part of your visions.

πŸ“

Location

Hyderabad, Telangana

πŸ“§
πŸ™
🐦

Twitter/X

x.com/itsZafeer (Most Active)