Sanjay P.

Sr. DevOps Engineer | AWS, Kubernetes, Terraform, CI/CD | CloudWatch

Your infrastructure works right up until the one person who understands it takes a week off. I fix that, permanently, and in code. | 8+ Years Of Experience I build and run production AWS infrastructure: Kubernetes and ECS, Terraform-managed environments, and CI/CD pipelines for teams that need to ship daily without hiring a full-time platform engineer. ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ WHAT PEOPLE HIRE ME FOR "Deploys are manual and everyone holds their breath." → GitHub Actions or GitLab CI pipelines with automated tests, staged rollouts and one-click rollback. Release stops being an event on the calendar. "The AWS bill keeps growing and nobody can explain it." → Cost audit, right-sizing, Savings Plans, scheduled shutdown of non-production, spot for anything interruptible, and tagging so next month's bill reads clearly without me. "All of our infrastructure lives in one engineer's head." → Everything into Terraform, remote state, reusable modules, plans reviewed in pull requests. Infrastructure becomes a repo instead of tribal knowledge. "We need containers but we can't babysit a cluster." → Production ECS or EKS with autoscaling, ingress, secrets management, GitOps through ArgoCD, and runbooks a tired engineer can follow at 2am. "We find out about outages from customers." → Prometheus, Grafana, Loki and alert rules tuned so a page means a human is genuinely needed. Fewer alerts, each one real. "A customer just sent us a security questionnaire." → Least-privilege IAM, private networking, encrypted state and secrets, hardened images, and an evidence trail that survives an auditor. ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ SOMETHING I BRING THAT MOST INFRASTRUCTURE PEOPLE DON'T I spent years as an application engineer before this. I know what's inside the container, not just how to schedule it, which means I debug the deploy that fails for application reasons rather than handing it back to your developers with a shrug. ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ STACK Cloud - AWS (EKS, ECS, EC2, RDS, ElastiCache, S3, CloudFront, Lambda, VPC, IAM, CloudWatch) IaC - Terraform, Ansible, CloudFormation, Helm Containers - Docker, Kubernetes, ECS, ArgoCD CI/CD - GitHub Actions, GitLab CI, Jenkins, Bitbucket Pipelines Databases - MySQL, PostgreSQL, Redis Observability - Prometheus, Grafana, Loki, ELK, CloudWatch Security - IAM, Secrets Manager, Vault, Trivy, WAF Scripting - Bash, Python ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ HOW I WORK 1. A 30-minute call. What are you running, how often do you deploy, and what broke most recently. 2. A written scope before anything starts, deliverables, timeline, price. 3. Work in the open. Shared channel, changes in reviewable PRs, a short demo each week. 4. Handover. Architecture diagram, runbooks, a walkthrough, so your team can change things without me. ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ My Upwork history shows one long application-development contract, 1,400 hours with a single client. The infrastructure work has been the larger part of my career and most of it happened off this platform. Judge it on the portfolio below and on a call. Tell me what you're running and what's hurting. You'll get a straight answer about whether I can help, including when the answer is no.