Prem Kumar

I build production-grade CI/CD pipelines that connect Jenkins, Docker, Terraform, and Ansible—then back it with Prometheus/Grafana observability. In one git push, my pipeline runs a SonarQube scan, parallel Docker builds, DockerHub pushes, and Ansible deployment.

On the infrastructure side, I provision AWS networking and compute with Terraform (VPC, EC2, ALB, Security Groups) and use Terraform remote state for safer team collaboration. I also automate EC2 deployments with Ansible playbooks structured around roles.

For Kubernetes operations, I deploy to kind and set up autoscaling with HPA plus RBAC-scoped monitoring discovery. I’ve worked on reliability details like PodDisruptionBudget and verified zero-downtime rollout/rollback behavior.

I’ve also extended DevOps into AI-assisted operations by building an LLM-powered Kubernetes diagnostic agent (FastAPI + Next.js) that streams telemetry for root-cause analysis, with an LLM + deterministic fallback path to keep diagnostics available during API outages.