Google Cloud Professional Cloud Developer tests whether you can build, deploy and run secure, scalable applications on Google Cloud, mainly on Cloud Run and GKE. The current exam guide (live since 24 April 2026) was restructured into four sections. It now includes generative AI APIs and AI-powered development tools: AI coding assistants, MCP servers in your IDE, AI-written unit tests and AI-assisted observability.
This guide follows Google's official exam guide: the four sections with weights, the services to know, a 6-week plan and 15 original scenario questions.
Quick facts (Google Cloud, November 2026)
| Item | Detail |
|---|---|
| Length | 2 hours |
| Questions | 50–60, multiple choice and multiple select |
| Fee | USD 200 (plus tax where applicable) |
| Languages | English, Japanese |
| Delivery | Online-proctored or at a test centre |
| Prerequisites | None; 3+ years of industry experience including 1+ year designing and managing Google Cloud solutions recommended |
Read the official Professional Cloud Developer page and exam guide before booking. Google also publishes free sample questions.
Who should take it?
- Backend and full-stack developers deploying containers on Cloud Run or GKE.
- Developers who passed Associate Cloud Engineer and want a developer-focused professional credential.
- AWS developers adding Google Cloud. Compare with AWS Developer – Associate.
Sections and weights
| Section | Weight |
|---|---|
| 1. Designing highly scalable, secure, and reliable cloud-native applications | ~32% |
| 2. Building and testing applications | ~23% |
| 3. Configuring cloud-native applications for deployment | ~24% |
| 4. Integrating applications with Google Cloud services | ~21% |
Section 1: Design (~32%)
- High-performing apps and APIs: Compute Engine vs GKE vs Cloud Run; regional vs zonal services; load balancers and session affinity; Memorystore caching; REST and gRPC APIs; Apigee and API Gateway for rate limiting and auth; Eventarc and Pub/Sub; traffic splitting for gradual rollouts and A/B tests; orchestration with Workflows, Cloud Tasks and Cloud Scheduler.
- Secure apps: Cloud Storage lifecycle and retention lock; Identity-Aware Proxy and Web Security Scanner; Artifact Analysis and Security Command Center findings; Secret Manager, Cloud KMS and Workload Identity Federation; Application Default Credentials, OAuth 2.0, Cloud SQL and AlloyDB Auth Proxy; least-privilege service accounts; Cloud Service Mesh, network policies and Direct VPC egress; Binary Authorization.
- Data: choosing storage; schemas for AlloyDB and Spanner vs Bigtable and Firestore; eventual vs strong consistency; signed URLs; writing to BigQuery.
| Need | Usual choice |
|---|---|
| Stateless HTTP container, scale to zero | Cloud Run |
| Many microservices needing Kubernetes control | GKE |
| Global relational database with strong consistency | Spanner |
| PostgreSQL-compatible with high performance | AlloyDB |
| Mobile app documents with offline sync | Firestore |
| Huge time-series or IoT writes | Bigtable |
| Temporary download link for a private file | Cloud Storage signed URL |
Section 2: Build and test (~23%)
- Local development with emulators from the gcloud CLI; Cloud Code, Cloud Shell, Cloud Workstations and Gemini Cloud Assist; IDEs with AI coding assistants and MCP servers.
- Cloud Build and Artifact Registry; build provenance for Binary Authorization.
- Unit tests written with AI coding assistants; automated integration tests in Cloud Build.
# cloudbuild.yaml: test, build, push, deploy to Cloud Run
steps:
- name: python:3.12
entrypoint: bash
args: ["-c", "pip install -r requirements.txt && pytest -q"]
- name: gcr.io/cloud-builders/docker
args: ["build", "-t", "asia-south1-docker.pkg.dev/$PROJECT_ID/apps/orders:$SHORT_SHA", "."]
- name: gcr.io/cloud-builders/docker
args: ["push", "asia-south1-docker.pkg.dev/$PROJECT_ID/apps/orders:$SHORT_SHA"]
- name: gcr.io/google.com/cloudsdktool/cloud-sdk
entrypoint: gcloud
args: ["run", "deploy", "orders", "--image", "asia-south1-docker.pkg.dev/$PROJECT_ID/apps/orders:$SHORT_SHA",
"--region", "asia-south1", "--no-traffic", "--tag", "canary"]Section 3: Deploy (~24%)
- Cloud Run: deploy from source; trigger with Eventarc and Pub/Sub; event receivers; version, expose and secure APIs with Apigee.
- GKE: deploy containers; liveness and readiness probes; Horizontal Pod Autoscaler.
Section 4: Integrate (~21%)
- Connections to Cloud SQL, Firestore and Cloud Storage; publishing and consuming Pub/Sub messages.
- Calling Google Cloud APIs with client libraries, REST, gRPC or API Explorer: batching, field masks to restrict returned data, pagination, caching and exponential backoff; using service accounts.
- Observability: metrics, logs and traces in Google Cloud Observability; Error Reporting; trace IDs across services; AI-assisted observability.
6-week study plan
| Week | Focus | Lab |
|---|---|---|
| 1 | Cloud Run, Artifact Registry, Cloud Build | Deploy a Python API from source; add a Cloud Build pipeline |
| 2 | GKE: deployments, probes, HPA, Workload Identity | Run the same API on GKE Autopilot with probes and HPA |
| 3 | Pub/Sub, Eventarc, Workflows, Cloud Tasks, Scheduler | Upload → Eventarc → Cloud Run thumbnail service |
| 4 | Firestore, Cloud SQL, AlloyDB, Spanner, Bigtable, Memorystore, signed URLs | Add Firestore storage and a Memorystore cache |
| 5 | Security: Secret Manager, KMS, IAP, Binary Authorization, least privilege | Move secrets to Secret Manager; protect an admin UI with IAP |
| 6 | Observability, Error Reporting, tracing; Gemini Cloud Assist; sample questions | Trace a slow request across two services; book the exam |
15 scenario questions with answers
These are original scenarios written for this guide, not exam questions.
1. A stateless API has spiky traffic and should cost nothing when idle. Where do you run it?
Cloud Run.
2. A new Cloud Run revision should get 5% of traffic first. How?
Deploy with no traffic, then split traffic between revisions (5% to the new one) and increase gradually.
3. An image upload to Cloud Storage must trigger processing. What connects them?
Eventarc (or a Pub/Sub notification) triggering a Cloud Run service.
4. A job must call an external API at a limited rate, with retries. What service?
Cloud Tasks.
5. An API key is hardcoded in source code. Where should it live?
Secret Manager, accessed by a least-privilege service account.
6. A GitHub Actions pipeline must deploy to Google Cloud without a service account key file. How?
Workload Identity Federation.
7. Only images built by your trusted pipeline may run on GKE. What enforces this?
Binary Authorization with build provenance from Cloud Build.
8. An internal admin web app must be reachable only by signed-in employees without a VPN. What do you use?
Identity-Aware Proxy (IAP).
9. Users need a 15-minute download link for a private report. How?
A Cloud Storage signed URL.
10. GKE pods receive traffic before the app has finished starting. What fixes it?
A readiness probe (and a startup probe for slow starts).
11. An API call returns HTTP 429 or 503 under load. How should the client react?
Retry with exponential backoff and jitter.
12. A list API returns huge responses and is slow. What helps?
Pagination and field masks to return only the fields you need, plus caching.
13. A request is slow somewhere across three microservices. How do you find where?
Distributed tracing with Cloud Trace, propagating the trace ID across services.
14. You need a public API with quotas, API keys and analytics for partners. What service?
Apigee (or API Gateway for simpler needs).
15. Repeated product lookups overload Cloud SQL. What do you add?
A Memorystore (Redis) cache.
After Professional Cloud Developer
- Practise with GKE interview questions and GCP interview questions.
- Read GCP for AI engineers and the Professional Cloud Architect guide.
- See live cloud developer jobs.
Cloud Soft Solutions in Ameerpet, Hyderabad teaches Google Cloud, GKE, DevOps and generative AI with real-time projects. See the APEX program and all courses.
📞 Call +91 96660 19191 · WhatsApp · Flat No. 513, 5th Floor, Nilagiri Block, Aditya Enclave, Ameerpet, Hyderabad
Frequently Asked Questions
How long is the Professional Cloud Developer exam?
2 hours, with 50 to 60 multiple-choice and multiple-select questions.
How much does the Google Cloud Developer exam cost?
USD 200, plus tax where applicable.
Which section has the most weight?
Designing highly scalable, secure, and reliable cloud-native applications, at about 32%.
Is there a prerequisite?
No. Google recommends 3+ years of industry experience, including 1+ year designing and managing solutions on Google Cloud.
Does the exam cover AI tools?
Yes. The current guide includes Gemini Cloud Assist, AI coding assistants and MCP servers in the IDE, AI-assisted unit tests and AI-assisted observability.
Which languages is the exam offered in?
English and Japanese.
