中文 | About Me | Resume | Blog

Projects

US Medical Resource Logistics System (Siglus)

When: 2022.07 - 2022.10
Stack: AWS, EFK, Terraform, Java, Docker, PostgreSQL

This project supported a healthcare logistics system aimed at helping an African country build a medical supply chain infrastructure.

My Contributions

  • Configured AWS Lambda with Terraform to automate scheduled EC2 start/stop and reduce waste
  • Built Prometheus-based monitoring and alerting, created Grafana dashboards, and forwarded alerts to WeCom through Flask
  • Used Ansible to manage test VMs and helped assemble deployment packages
  • Independently explored SkyWalking integration to improve monitoring and performance analysis

Singapore Swab Registration System (SRS)

When: 2022.10 - 2024.12
Stack: Azure, NestJS, Next.js, Docker, Kubernetes, Ansible, Istio, Vault, Terraform

This Singapore public-health system supported COVID-19 ART/PCR appointment booking and result submission.

My Contributions

  • Migrated CI/CD from Bamboo to GitLab CI and optimized parent-child pipelines for a monorepo
  • Maintained and upgraded deployment tooling with Ansible
  • Resolved a long-standing Terraform state synchronization issue shortly after joining
  • Managed migration from Azure Single Server to Azure Flexible Server, including schema and networking challenges
  • Operated services with Istio to strengthen service reliability and security
  • Mentored junior teammates on Terraform, Kubernetes, and CI/CD practices during project iterations

Singapore Children’s Health E-Services

When: 2024.08 - Present
Stack: Azure, Node.js, Docker, Kubernetes, Ansible, Helm

This project focuses on digital health services for children, serving both parents and public health organizations.

My Contributions

  • Served as the sole DevOps engineer, independently managing cloud ops, tool upgrades, monitoring, and CI/CD
  • Participated in infrastructure rebuild and cloud migration, taking on stronger coordination responsibility during team lead absence
  • Handled production dependency issues, made fast recovery decisions, and strengthened monitoring and fault tolerance
  • Improved deployment workflows with Helm and refined project monitoring metrics