Google Cloud Skills Summary | Generated by AI

Home 2026.07

Based on all the notes I’ve read, here’s a comprehensive summary of your Google Cloud knowledge:

☁️ Google Cloud — What You Know & Have Used

🛠️ Actually Built & Deployed (Hands-on)

Module/Feature How You Used It
Speech-to-Text v2 API Built into your ww CLI tool — ww gcp-speech transcribe <audio> — uploads audio to GCS, runs Chirp/long/short models, saves transcripts as .md. You know Chirp, Chirp_2, Chirp_3, long, short models in detail.
Cloud Storage (GCS) The test2x bucket for audio uploads. GCS lifecycle rules (delete old files, move to Nearline/Coldline/Archive). gsutil CLI.
Cloud Run Deployed a Java/Spring Boot blog-server container. You know gcloud run deploy --source ., Artifact Registry, Cloud Build, .gcloudignore, --verbosity=debug for debugging hangs. Deployed to asia-northeast1 region.
YouTube Data API v3 Built ww gen-video upload — OAuth 2.0 desktop flow, video upload with resumable MediaFileUpload. Stored client_secret.json at ~/.google/.
OAuth 2.0 Fixed redirect_uri issues, ran flow.run_local_server(port=8080), handled token caching at ~/.google/youtube_token.json.
Compute Engine Created E2 micro VM in Taipei (asia-east1) as a VPN server (around $13/month). Knows shared-core, preemptible/spot VMs, custom machine types.
gcloud CLI Updated from 507.0.0 → 532.0.0. Uses gcloud auth login, gcloud config set project, gcloud services enable, gcloud components update.
Google AdSense Added to your Jekyll blog — ads.txt, Auto Ads script, manual in-article ad units. Applied for approval, dealt with rejection/fixes. Understands ad revenue dynamics.

📚 Studied & Know Deeply (Exam/Cert Level)

Module What You Know
BigQuery Serverless columnar warehouse. Dremel/Colossus/Borg architecture. Partitioning & clustering for cost optimization. bq CLI, Python client (google.cloud.bigquery), --dry_run. BigQuery ML, vector search, JSON embeddings storage for RAG.
IAM & Security Identity-Aware Proxy (IAP) — Zero Trust, context-aware access. roles/compute.instanceAdmin vs roles/editor vs roles/owner. Service accounts vs user accounts.
GKE (Kubernetes Engine) Knows GKE cluster creation, HPA/VPA scaling, Nginx Ingress, secure access patterns.
Cloud SQL / Spanner / Firestore Spanner — globally distributed, strong consistency, ACID. Firestore — NoSQL document DB, real-time sync, offline support, security rules. Cloud SQL — traditional relational.
Vertex AI Imagen 4 image generation on Vertex AI, safety filters, google-gen-ai-image-config. Google AI Studio vs Vertex AI metrics.
Cloud Load Balancing Introduction to GCP Cloud Load Balancing, Cloud NAT, Cloud Armor (WAF).
Networking Cloud NAT vs home router, multi-region design, VPC, private service connect.
Certifications Studied GCP ACE (Associate Cloud Engineer), Professional Cloud Developer, Cloud Digital Leader, Professional ML Engineer. Did practice questions.

💰 Pricing & Cost Knowledge

Topic Details
Compute Engine pricing E2 micro = $0.0084/hr in Taipei. Sustained use discounts, committed use discounts (30-70% off). Spot VMs (60-91% off).
Network egress $0.12/GiB from Asia (first 1TB). Ingress is free.
GCS storage classes Standard → Nearline → Coldline → Archive. Lifecycle rules.
BigQuery pricing $6.25/TB scanned (on-demand). Partitioning/clustering to reduce.
GPU costs 8× H100 on GCP = ~$57k/month. GCP GPU quota denials experienced.
Cloud Run Free tier: 2M requests/month. Pay-per-request.

⚡ Summary of Your GCP Profile

You’re not just studying theory — you’ve built real integrations:

You know the developer/platform side of GCP well (Cloud Run, GCS, Speech API, YouTube API, gcloud CLI, BigQuery), plus certification-level knowledge across many services. Your weakest areas (based on note volume) seem to be GKE (more theory than hands-on) and Vertex AI (studied Imagen but less hands-on), while your strongest is GCP Speech-to-Text + Cloud Run + GCS + OAuth/YouTube integration.

Generated by AI. Curating and sharing still takes effort. If you find it useful, feel free to donate. WeChat: @lzwjavaWeChat QR · X: @lzwjava · Say hi 👋

Back Donate