Projects

PROBLEM

Production Kubernetes environments need clear reliability signals before incidents become customer-facing. The work focused on making service health visible through practical SLOs, dashboards, and alerting paths.

SOLUTION

Built observability coverage around Kubernetes workloads using Datadog, OpenTelemetry, and Terraform-managed configuration so reliability signals could be reviewed and changed as code.

OUTCOME

Improved incident visibility, reduced manual investigation work, and gave engineering teams a shared vocabulary for service health, error budgets, and operational follow-up.

PROBLEM

Manual deployment steps and inconsistent release checks make production changes harder to reason about, especially when teams need repeatable rollouts and quick recovery paths.

SOLUTION

Worked on deployment automation with GitHub Actions, Kubernetes rollout patterns, and Terraform guardrails to make release behaviour more predictable.

OUTCOME

Reduced toil around releases and made deployment state easier to inspect, repeat, and recover from when production changes needed attention.

PROBLEM

Repeated infrastructure work becomes fragile when teams rely on manual setup, local scripts, and one-off cloud configuration that is hard to review or reproduce.

SOLUTION

Used Terraform, Go, and shell automation to standardise common infrastructure workflows and keep cloud changes reviewable through source-controlled definitions.

OUTCOME

Improved repeatability for infrastructure changes and reduced the amount of context needed to understand, review, and operate common platform tasks.