DevOps engineer working across individual contribution and technical leadership, responsible for cloud infrastructure decisions as well as hands-on implementation. Core areas: cloud migrations, systems administration, cloud architecture design, cost optimization, and CI/CD automation, built on Terraform and GitHub Actions.
A platform shouldn't stop at "it deploys." It should carry infrastructure from a Terraform plan to a monitored, cost-aware production service through a controlled path:
Provision -> Build -> Deploy -> Observe -> Optimize
- Provision: VPCs, ECS/Fargate, RDS, and IAM defined through versioned Terraform modules pulled from a private registry, across a multi-account AWS Organizations setup with centralized SSO.
- Build: Services built through Docker Bake and matrix/multi-platform GitHub Actions pipelines, running in parallel instead of one long serial job.
- Deploy: GitHub Actions ships through ECR to ECS Fargate, with blue-green/canary rollouts, environment-gated approvals from staging to production, and automated rollback on failed health checks.
- Observe: Prometheus and Grafana cover the stack, with CloudWatch as the first place I check when something breaks.
- Optimize: RDS Reserved Instances, Savings Plans, and right-sized Fargate tasks keep a recurring FinOps pass grounded in real usage.
| 🏗️ Infra as code | 100% Terraform, private module registry, no console click-ops |
| 🚀 Deploy pipeline | GitHub Actions -> Docker Bake -> ECR -> ECS Fargate |
| 🔐 Security posture | Least-privilege IAM, layered Security Groups, Gitleaks scanning on every repo |
| 📈 Observability | Prometheus + Grafana on ECS/EFS, unmonitored infra doesn't count as production to me |
| 💰 Cost discipline | RDS Reserved Instances, Savings Plans, right-sized Fargate tasks |
| 🏢 AWS accounts | Multi-account landing zone via AWS Organizations, centralized SSO |
- Migrated on-prem/VPS workloads to AWS and broke a monolith into containerized microservices running on ECS/Fargate
- Retired legacy pipelines in favor of GitHub Actions, with reusable workflows shared across repos
- Built blue-green/canary deployment flows with environment-gated approvals (staging -> production) and automated rollback on failed health checks
- Set up matrix/multi-platform GitHub Actions builds and Docker Bake multi-target builds running in parallel
- Stood up a multi-account AWS Organization with centralized SSO
- Maintain a private Terraform module registry so infra changes stay reviewable, versioned, and consistent across environments
Cloud & Infrastructure:
CI/CD & Delivery:
Security & Governance:
Observability & Cost:
Languages & Automation:
A small vineyard, the stray cats around it, and a telescope out whenever the sky's clear. R.I.P. Sufle, the calmest cat in the vineyard 🐈