Lead Platform Engineer / Solution Architect - Google Cloud (GCP)
A hands-on architect/engineer role with a Google Cloud Partner in Sydney (hybrid, 3 days in office), leading large customers' migrations off VMware and other clouds onto GCP through to go-live. Core stack: GCP compute (GKE, Cloud Run, serverless), Terraform/GitOps, plus Vertex AI and BigQuery for AI/ML workloads.
- Compensation
- A$170K–A$210K / yr
- Location
- Sydney NSW
- Work setup
- Hybrid
- Seniority
- Lead
- Employment
- Not specified
- Experience
- Not specified
- Category
- Devops
- Posted
Skills
Must have
- Commercial experience architecting and shipping on Google Cloud
- Depth across GCP Compute, Serverless, GKE, Cloud Run, networking, and data services
- Terraform and GitOps fluency at a level clients can maintain
- Comfort with Python, Go, or Bash
- Consulting instincts: run requirements sessions and defend proposals unscripted
- Track record of building things hands-on, not just approving others' builds
Nice to have
- Experience with Vertex AI or BigQuery pipelines
- Google Professional Cloud Architect certification (or equivalent hands-on experience)
Full description
Salary: $170,000 – $210,000 per year
The fun part first... ☁️💡kubectl apply --dry-run: hire_principal_cloud_architect$ kubectl apply -f principal-architect.yaml --dry-run=server
architect.roles/principal-cloud-architect configured (dry run)
+ spec.title: "Principal Google Cloud Architect"
+ spec.location: "Hybrid, 3 days"
+ spec.reportsTo: "Head of Cloud Practice"
+ spec.status: "SCALING"
+ status.legacyMigrations: "in-progress"
+ status.clientTrust: "high"
+ status.slideDeckArchitecture: "destroy"
deployment strategy: RollingUpdate
1 role to create, several legacy diagrams to retire, 1 practice to scale!
The honest version of this ad:
Most Principal Engineer and Solution Architect ads read like a certification checklist wearing a blazer. We're not doing that.
Here's what's actually true: We're supporting a Google Cloud Partner working with genuinely large, complex customers, the kind still mid-migration off VMware and other clouds, not the kind who “just need someone to tidy up the diagrams.” You'd be the person customers trust to turn a messy Rehost / Replatform / Refactor decision into an architecture that actually survives contact with production.
If you’ve ever sat in a stakeholder meeting translating "why is this so slow" into a real GCP fix, keep reading!
What you'd actually be doing:
- Designing the migration, not just diagramming it: Leading Rehost, Replatform, and Refactor strategies off VMware and other clouds onto GCP, and staying hands-on through to go-live, not handing it off after the pitch deck.
- Running Compute, Serverless, GKE, and Cloud Run like they're all in your muscle memory: Picking the right compute model per workload instead of defaulting to whatever's familiar.
- Making IaC and GitOps the default, not the exception: Terraform modules and GitOps pipelines that clients can actually own after you leave the room.
- Bringing AI/ML into the architecture conversation: Vertex AI and BigQuery pipelines, with a working knowledge of where agentic frameworks (Agent Builder, ADK) genuinely fit vs. where they're hype.
- Being the "why" behind the "what": Translating gnarly technical trade-offs into a narrative an exec will actually approve.
- Mentoring as you go: Levelling up the Engineers around you, not hoarding the interesting problems.