Cloud & Data Infrastructure
A production-shaped AWS environment built end to end: analytics collection, a lakehouse, CI/CD, a Kubernetes ML platform, and the security posture holding it together. All of it Terraform-deployed and reproducible.
How this site is stored, encrypted, and served from the edge — and the alarms watching it.
Architecture →Behavioral analytics pipeline with a custom event schema, a private Iglu registry, and five enrichments.
Architecture →Lakehouse behind PrivateLink with customer-managed keys, plus dbt as the transformation layer.
Architecture →EKS wrapped in an Istio service mesh with mTLS, the Prometheus/Grafana stack, and GitOps delivery.
Architecture →The Jenkins build server and the delivery model behind it — Terraform plan/apply pipelines and the dbt docs build.
Architecture →Distributed compute, interactive notebooks, and experiment tracking running on the cluster.
Architecture →Live when the stack is up.
| Service | What it does | Link |
|---|---|---|
| Jenkins | CI/CD server — see DevOps | jenkins.patrick-cloud.com |
| Snowplow Collector | Event collection endpoint | collector |
| Snowplow Iglu | Schema registry | iglu |
| Grafana | Metrics dashboards | /grafana |
| Prometheus | Metrics collection | /prometheus |
| Kiali | Service mesh console | /kiali |
| Argo CD | GitOps delivery | /argocd |
| dbt Docs | Model documentation | /dbt-docs |
| Ray Dashboard | Cluster compute | /ray-cluster-dashboard |
| Ray Serve | Model serving endpoint | /ray-service-serve |
| Jupyter | Notebooks | /jupyter |