Cloud/DevOps Engineer who obsesses over zero-downtime deployments and keeping infrastructure invisible to developers. ASU Computer Science graduate, bilingual (English/Korean), U.S. Green Card holder.
Currently studying for my AWS Certified Solutions Architect – Associate and looking for Junior/Associate/Entry-level Cloud or DevOps opportunities.
Contributed to a 3-phase infrastructure evolution (MVP CRUD → ECS CI/CD → EKS GitOps) for a 5-microservice system as part of a 4-person team. After self-studying ArgoCD and Argo Rollouts on a mentor's suggestion, I built and delivered the team's full Blue-Green deployment system — team awarded Best Team (1st out of 4 teams).
- Eliminated 2–3 min downtime by migrating from ECS Rolling Update to EKS + Argo Rollouts Blue-Green strategy; achieved 100% availability across 5 microservices
- Reduced build time 68% (25 min → 8 min) via Gradle dependency caching + Docker layer caching on ECR buildcache; cut image size 66% with multi-stage builds
- Cut MTTR 90% by designing Prometheus metric-based Analysis Templates for auto-rollback, with real-time Discord alerts for incident visibility
- Designed an ArgoCD App of Apps pattern for hierarchical GitOps — single-command dev/prod sync across all K8s resources
Stack: AWS EKS · ECR · RDS · ElastiCache · MSK · Kubernetes · Helm · Docker · GitHub Actions · ArgoCD · Argo Rollouts · Prometheus · Grafana · Linkerd
Capstone project for a startup neobank serving U.S. military veterans, working directly with the company's CEO and CTO as project sponsors. Contributed to an automated document-verification pipeline on AWS (event-driven, replacing a manual military-ID review process) and co-authored the system design documentation.
Details withheld — project was covered under an NDA with the sponsor company.
Designed and built a serverless, event-driven backend architecture (API Gateway → Lambda → SNS/SQS) with fully modularized infrastructure-as-code.
- Cut infrastructure costs 70% with async event-driven architecture
- Reduced provisioning time 99% (2 days → 20 minutes) via Terraform IaC modularization (VPC, ECS, RDS, Lambda)
- Achieved zero downtime incidents (previously 5/day) via ALB + ECS Auto Scaling
- Built serverless DynamoDB transaction validation for automated payment confirm/cancel flows; automated CI/CD via GitHub Actions
Stack: Node.js · Terraform · Docker · AWS ECS/Fargate/Lambda/RDS/DynamoDB/SNS/SQS/SES/API Gateway · GitHub Actions
| Category | Tools |
|---|---|
| Infrastructure | AWS (EKS, ECR, RDS, ElastiCache, MSK, Lambda, CloudFormation, Amplify), Terraform, Helm |
| Container & Orchestration | Docker, Kubernetes, EKS, K3d, Linkerd (Service Mesh) |
| CI/CD & GitOps | GitHub Actions, ArgoCD, Argo Rollouts |
| Observability | Prometheus, Grafana, CloudWatch |
| Languages | Python, Java, C++, Bash, Node.js |