Practical platform engineering

Projects

Project work framed like production preparation: validate failure behavior, deployment workflows, runner capacity, backup assumptions, observability coverage, and operator runbooks before trusting a pattern in a real environment.

3 total projects 2 active See experience context

Project areas

CI/CD infrastructure GitOps delivery Kubernetes reliability Observability Rollout safety Runner reliability

Tools and skills

ansible argo-rollouts argocd docker gitlab grafana jaeger k3s kubernetes logging prometheus terraform

What fails first: resource pressure, rollout state, runner saturation, storage, or signal quality.

How operators detect it: metrics, logs, traces, events, dashboards, and alert behavior.

How change is controlled: Helm values, GitOps sync, pipeline gates, and rollback criteria.

What remains documented: runbooks, incident notes, recovery assumptions, and follow-up work.

Kubernetes Observability Project

active

A Kubernetes observability project for testing whether Prometheus metrics, Grafana dashboards, Jaeger traces, logs, and alert rules help an operator isolate real failure modes.

k3s prometheus grafana jaeger logging terraform

GitLab Runner Reliability Project

active

A GitLab CI/CD runner project for separating pipeline design problems from runner saturation, with focus on queue behavior, cache correctness, executor isolation, and retry discipline.

gitlab docker terraform ansible

Argo CD Rollout Guardrails

planned

A GitOps delivery project for testing Helm values contracts, Argo CD sync behavior, rollout health checks, promotion gates, and rollback runbooks before standardizing a release pattern.

argocd argo-rollouts prometheus kubernetes