Achilleas Triantafyllou
I keep global infrastructure boring.
Senior Infrastructure Engineer. I build cost-efficient, multi-region real-time infrastructure that lets the business keep growing well past the usual scaling walls. Currently @ gather.town. Previously @ Citrix.
What I do
- Reliability
- Scale & cost
- Observability
- Incident response
- Platform engineering
- CI/CD
- Architecture & scoping
How I work
Infra engineer and SRE, driven by product needs and real problems. I measure my work in operational excellence — runbooks a sleepy on-call can follow, dashboards that point at the thing that broke, and error messages that tell you what to do next.
Experience
-
Sept 2021 — nowSenior Infrastructure Engineer
-
2019 — 2021Platform EngineerCitrix
-
MScElectrical Engineering & Computer ScienceUniversity of Patras
Stack
- Coding
-
preferable choice Golang · Python
picks up whatever else the job needs - Infra-as-code
- Terraform · Terragrunt · Atlantis · Ansible
- Cloud & edge
- AWS · Azure · GCP · DigitalOcean · Cloudflare
- Kubernetes
- EKS · AKS · DOKS · Helm
- Ingress & proxies
- ingress-nginx · Traefik · HAProxy · aws-load-balancer-controller
- CI/CD
- CircleCI · GitHub Actions · ArgoCD · Jenkins
- Observability
- Prometheus · Grafana · Alertmanager · New Relic · Elasticsearch · OpenTelemetry · Datadog · Splunk
- On-call & ops
- incident.io · PagerDuty
Selected work
- Infrastructure cost reduction without breaking SLOs — density instrumentation, iterative fleet scale-downs, and autoscaling for stateful real-time services.
- Platform architecture — multi-region topology, ingress strategy, and the design docs that set the team's direction for the next quarter.
- Multi-region Kubernetes version upgrades, with the companion stack (autoscaler, CNI, ingress, observability) upgraded in lockstep.
- Observability pipeline rebuilds with per-environment routing and clean data flow across clusters.
- Custom Kubernetes controllers that close gaps in managed-service behavior — cluster lifecycle, upgrade orchestration, and the thin layers that make operations safer.
- Long-tail reliability work — memory and resource leaks, plus incident response hardening.