Building reliable
AWS & Kubernetes platforms.

Helped move legacy deployment patterns toward a more scalable Kubernetes-based model using Helm, Terraform, and ArgoCD, improving consistency across environments and creating a stronger foundation for repeatable releases.

Reduced deployment time by around 70% through pipeline optimization, better automation, and cleaner release flow design, helping engineering teams ship changes faster with fewer manual steps and better confidence in the process.

Reduced AWS staging infrastructure cost by roughly 46% through usage review, smarter instance scheduling, and resource right-sizing, showing that reliability and cost awareness can be improved together instead of treated as separate priorities.

Rolled out Prometheus and Grafana based monitoring to provide centralized visibility into service health, performance, and operational signals, helping teams detect issues earlier and reduce time spent debugging production behavior.

We start with a free 20-minute call. Then I audit your setup — your AWS bill, your pipelines, or your cluster — and tell you honestly what's worth fixing.

You get a fixed-scope plan in writing: what I'll deliver, what it costs, and when it ships. No hourly meters, no scope creep.

I do the work async with regular updates — zero disruption to your team's day. You always know what's done and what's next.

Docs, dashboards, and a walkthrough — your team owns everything after I'm gone. No lock-in, no mystery boxes.

Leading core DevOps responsibilities across CI/CD, cloud infrastructure, Kubernetes adoption, and observability for multiple products, with strong emphasis on release reliability, delivery speed, and infrastructure efficiency.

• Migrated legacy workloads from EC2 to Kubernetes using Helm, ArgoCD, and Terraform.
• Optimized Jenkins pipelines and integrated automated tests, reducing deployment time by around 70%.
• Reduced AWS staging costs by roughly 46% through instance scheduling, usage audits, and right-sizing.
• Implemented Prometheus and Grafana for centralized monitoring and faster incident response.
• Worked across infrastructure and application delivery concerns to improve operational consistency between environments.

Built and maintained automation pipelines and cloud environments for containerized applications, helping teams standardize deployments and reduce manual operational effort.

• Created CI/CD workflows with GitHub Actions for build, test, and deployment stages.
• Developed Shell and Python automation scripts for backups, housekeeping, and log management.
• Deployed and managed containerized applications on AWS and Kubernetes.
• Supported development teams with infrastructure troubleshooting and smoother release workflows.

Built and maintained CI/CD pipelines and cloud infrastructure for containerized applications running on AWS and Kubernetes.

• Created CI/CD workflows with GitHub Actions for build, test & deploy.
• Developed Shell & Python automation scripts for backups and log management.
• Deployed and managed containerized apps on AWS & Kubernetes.

Led DevOps initiatives to improve scalability, reliability, deployment consistency, and operational efficiency for a major grants management product used in a production environment.

• Moved from EC2-based deployments to Kubernetes with Helm and ArgoCD.
• Redesigned staging infrastructure, cutting costs while keeping performance intact.
• Built Python scripts to generate daily test reports and speed up debugging cycles.
• Strengthened delivery workflows by making infrastructure and deployment steps more repeatable.

Introduced centralized monitoring and alerting patterns to reduce operational blind spots, improve mean time to resolution, and give teams a clearer picture of service behavior across environments.

• Rolled out Prometheus and Grafana template dashboards for key services.
• Defined SLIs around latency, error rate, and resource saturation.
• Helped teams debug issues faster by correlating logs, metrics, and deployments.
• Improved the quality of operational visibility so alerts could better support action instead of noise.

Cloud bills were growing faster than usage. I broke costs down service-by-service, found the real drivers — data transfer, idle capacity, oversized instances — and delivered a prioritized savings plan with calculator-verified numbers. This is the same audit process I now offer as a fixed-scope freelance service.

• Bottom-up modeling across compute (EKS/EC2), databases, storage, and data transfer.
• Cross-checked estimates against live production billing data for accuracy.
• Identified the real cost drivers most teams miss: transfer, storage growth, and idle capacity.

Tired of 2 AM "CPU is high" alerts that say nothing, I built a monitoring system that thinks like an SRE: detects issues in real time, finds the root cause with AI, suggests fix steps, and auto-creates Jira tickets.

• Pipeline: Flask → Logstash → Elasticsearch → Kibana, with an AI monitor on top.
• 5-layer optimization cuts AI inference cost by ~90% — the right model at the right time.
• Alerts carry root cause + suggested fix + one-click ACK instead of noise.

View on GitHub →

Public repos where I practice what I preach — real infrastructure code, not tutorial copies:

• terraform-2tier-architecture — modular 2-tier AWS infrastructure (networking, app and data layers).
• TerraformAWSInfra — S3 + CloudFront static hosting with remote Terraform backend.

Validated my ability to design and operate scalable, secure, resilient, and cost-optimized solutions on AWS, reinforcing the architectural side of my hands-on DevOps and platform work.

Completed multiple learning programs across containers, CI/CD, and cloud fundamentals, strengthening the practical knowledge I apply in day-to-day infrastructure and delivery work.