Samsung Electronics
From alerts to visibility.
Built Kubernetes resource monitoring and automation, then extended the work into a resource-usage dashboard.
At a glance
The story
The problem
Underused namespaces were holding cluster resources, and there was no easy way to see how namespace usage changed over time.
The scope
My internship began with resource monitoring and automation for Kubernetes. The scope grew into a dashboard that made historical namespace usage and administrative activity easier to inspect.
The build
I used Kubernetes, Prometheus, Flask, and PostgreSQL for the monitoring and dashboard work. I implemented email alerts with Kubernetes CronJob and Knox API, and used Jenkins for deployment automation.
The result
The project reclaimed 2,000 vCore CPU and 6,000 GiB memory from underused namespaces.
What came next
I received a full-time offer and chose to continue toward a master's degree. I returned to industry for an Apple internship the following summer.