gcp-expert

Google Cloud Platform expertise from Google engineers. Master GCP services, GKE, BigQuery, Vertex AI, and cloud-native architecture. Use when working with Google Cloud services, deploying to GCP, or d

SKILL.md

GitHub

GCP Expert - Google Cloud Platform Mastery

Experts: Emily Rodriguez (Google AI, 10 years), Alex Kim (Google SRE, 11 years) Level: 10/10 - Google's own cloud platform

Overview

Google Cloud Platform from Google engineers - the infrastructure powering Search, YouTube, Gmail, Maps. This is cloud expertise from the source.

When to Use GCP Services

Compute needs:

  • Containers/Kubernetes → GKE (see references/compute.md)
  • Serverless containers → Cloud Run (see references/compute.md)
  • VMs → Compute Engine (see references/compute.md)

Data & Analytics:

  • Data warehouse → BigQuery (see references/data.md)
  • Streaming pipelines → Dataflow (see references/data.md)

Databases:

  • Relational → Cloud SQL or Cloud Spanner (see references/databases.md)
  • NoSQL → Firestore (see references/databases.md)

AI & ML:

  • Machine learning → Vertex AI (see references/ai-ml.md)

Messaging:

  • Pub/Sub, event-driven → See references/messaging.md

Core Workflow

  1. Design: Choose appropriate services for requirements
  2. Provision: Use Terraform or gcloud CLI
  3. Deploy: GKE for containers, Cloud Run for serverless
  4. Monitor: Cloud Monitoring + Cloud Logging
  5. Optimize: Cost optimization + performance tuning

Best Practices

Security

  • IAM: Least privilege access
  • VPC Service Controls: Perimeter security
  • Secret Manager: Never hardcode secrets
  • Workload Identity: For GKE pods

Cost Optimization

  • Committed Use Discounts: Save 57% on VMs
  • Preemptible VMs: Save 80% for batch jobs
  • BigQuery slots: Reserve for predictable costs

Reliability

  • Multi-region for critical services
  • Health checks on all services
  • Auto-scaling for traffic spikes

Monitoring

  • Cloud Monitoring: Metrics + alerts
  • Cloud Logging: Centralized logs
  • Cloud Trace: Distributed tracing

Google's Production Scale

GCP powers:

  • Google Search: 8.5 billion searches/day
  • YouTube: 2 billion users, 1 billion hours/day
  • Gmail: 1.8 billion users
  • Google Maps: 1 billion users

Related Skills


Last Updated: 2026-02-03 Experts: Emily Rodriguez, Alex Kim (Google, 10-11 years each)