DevOps / Platform Engineer
I build cloud platforms and deployment pipelines. I use code to kill manual toil, scale infrastructure, and improve system reliability. Most of my work involves making sure rollbacks don't wake people up at 3 AM.
I prefer GitOps because I've seen what happens when clusters become the source of truth. Manual kubectl applies create drift. Forgotten config changes break production. So I use Terraform for infrastructure, Kubernetes for workloads, and CI/CD pipelines to catch broken YAML before it ships.
This portfolio demonstrates an Internal Developer Platform that cuts environment setup from hours to minutes. It automates infrastructure provisioning, handles deployment failures automatically, and stops "works on my machine" problems before they reach production.
A self-service platform that automates infrastructure and deployment workflows, allowing development teams to ship faster with confidence.
A simple, automated deployment flow
Simple form or API call with app configuration
Check resources, naming, and permissions
GitHub Actions workflow starts automatically
Terraform ensures cloud resources are ready
Containers are rolled out with health checks
Developer receives confirmation and app URL
Learning from production failures
Pod crash loop due to missing environment variable
Root cause: Configuration was not validated before deployment
Fix: Added pre-deployment validation step in pipeline
Manual changes made outside Terraform caused conflicts
Root cause: No enforcement of infrastructure-as-code policy
Fix: Implemented drift detection and automated reconciliation
Database migration prevented automatic rollback
Root cause: Migration ran before deployment health check
Fix: Separated migrations from deployments, added rollback-safe patterns
Key takeaway: Real platforms aren't about perfect code—they're about graceful failure handling, observability, and continuous improvement.