Cloud Engineer Fresher Roadmap: 12 Labs in 12 Weeks
What you build: a 12-week cloud engineer plan for freshers with one tested lab every week, plus a real Week 0 readiness check that proves billing, a budget, a free VM and a bucket work.

✍️ Translate this blog into Hindi, Tamil or Telugu
Most cloud roadmaps are long lists of things to "learn". This one is a 12-week plan where every week ends with a hands-on lab on Google Cloud that you really run, so that after three months you have done the work, not just read about it. It is made for freshers and career-changers with no cloud experience: about 6 to 8 hours a week, mostly on free tiers. Before Week 1 you run a real readiness check (in this post) that proves your account is ready: billing linked, a budget alert set, and a free VM and a bucket created and deleted by script.
Who it is for: students, freshers and career-changers aiming for a first cloud engineer, cloud support or junior DevOps role. Time: 12 weeks of about 6 to 8 hours. Cost: almost every lab runs on free tiers or the free trial credit; each lab post says what it costs.
The readiness check below was run for real by the CareerByteCode Team on a fresh Google Cloud project, and every output is the real one we got.
🎓⭐ Recommended by CareerByteCode · Live trainingAWS Cloud Architecture and AIOps Bootcampโน10,620See the training →Part A: how the roadmap works

Three rules make it work:
- One lab every week, finished and cleaned up. A lab is done when you can explain each step and you have deleted what you created.
- Write a short note after each lab (what you built, what broke, how you fixed it). By Week 12 these notes become your portfolio and your interview stories.
- Free first. Use Always Free products and the trial credit; set a budget alert before anything else (Week 0).

Real-life example: learning to swim. You do not read twelve books about water; you get into the pool every week, a little deeper each time, with a lifeguard nearby (the budget alert).
Week 0: the readiness check (the lab of this post)
Before Week 1, make sure your account is ready. Open Cloud Shell (the >_ icon at the top right of the Google Cloud console), set your project, and save this script as readiness.sh:
export PROJECT_ID="your-project-id"
gcloud config set project $PROJECT_ID
gcloud services enable compute.googleapis.com storage.googleapis.com#!/bin/bash
# readiness.sh: week 0 of the roadmap. Checks that your Google Cloud setup is ready for 12 weeks of labs,
# and proves it by creating (and deleting) a free VM and a bucket. Run it in Cloud Shell.
P=$(gcloud config get-value project 2>/dev/null)
ok() { echo "[OK] $1"; }; bad() { echo "[FIX] $1"; }
[ -n "$(gcloud config get-value account 2>/dev/null)" ] && ok "signed in to gcloud" || bad "run: gcloud auth login"
[ -n "$P" ] && ok "a project is selected" || bad "run: gcloud config set project <id>"
[ "$(gcloud billing projects describe $P --format='value(billingEnabled)' 2>/dev/null)" = True ] && ok "billing is linked (needed even for free tier)" || bad "link billing: Billing > Account management"
BA=$(gcloud billing projects describe $P --format='value(billingAccountName)' 2>/dev/null | sed 's#billingAccounts/##')
N=$(gcloud billing budgets list --billing-account=$BA --format='value(name)' 2>/dev/null | wc -l); [ "$N" -gt 0 ] && ok "a budget alert exists ($N found)" || bad "create a budget: Billing > Budgets & alerts"
for api in compute storage; do gcloud services list --enabled --format='value(config.name)' | grep -q "^$api.googleapis.com" && ok "$api API is on" || bad "gcloud services enable $api.googleapis.com"; done
S=$(date +%s); gcloud compute instances create ready-vm --zone=us-east1-b --machine-type=e2-micro --image-family=debian-12 --image-project=debian-cloud --boot-disk-size=10GB --boot-disk-type=pd-standard --no-service-account --no-scopes > /dev/null 2>&1 \
&& gcloud compute instances delete ready-vm --zone=us-east1-b --quiet > /dev/null 2>&1 && ok "a free e2-micro VM was created and deleted in $(( $(date +%s) - S )) s" || bad "VM test failed: check quotas and the Compute API"
B=ready-$RANDOM$RANDOM; gcloud storage buckets create gs://$B --location=us-east1 > /dev/null 2>&1 && echo hi | gcloud storage cp - gs://$B/hi.txt > /dev/null 2>&1 && gcloud storage rm -r gs://$B > /dev/null 2>&1 && ok "a bucket was created, written and deleted" || bad "bucket test failed"
echo "VMs left: $(gcloud compute instances list --filter='name=ready-vm' --format='value(name)' | wc -l), test buckets left: $(gcloud storage buckets list --format='value(name)' | grep -c '^ready-')"It checks that you are signed in, a project is selected, billing is linked, a budget exists and the two most used APIs are on. Then it proves it: it creates a free e2-micro VM and deletes it, and creates, writes and deletes a bucket, and finally counts what is left (should be 0).
Run it
# Week 0: run the readiness check (it creates and deletes a free VM and a bucket to prove everything works)
grep -c 'ok "' readiness.sh | sed 's/$/ checks in the script/'
chmod +x readiness.sh && bash readiness.sh7 checks in the script
[OK] signed in to gcloud
[OK] a project is selected
[OK] billing is linked (needed even for free tier)
[OK] a budget alert exists (1 found)
[OK] compute API is on
[OK] storage API is on
[OK] a free e2-micro VM was created and deleted in 162 s
[OK] a bucket was created, written and deleted
VMs left: 0, test buckets left: 0
All green: billing is linked, a budget alert exists, a free VM was created and deleted (about 2 to 3 minutes), and nothing is left behind.
Break it: a typo in the project id
The most common Week 0 problem: a project id that is slightly wrong (the id is not the project's display name; it is the unique id, often with numbers at the end).
# Break it: the most common beginner mistake, a typo in the project id. Run the check against "my-project-typo-123"
CLOUDSDK_CORE_PROJECT=my-project-typo-123 bash readiness.sh 2>&1 | head -6[OK] signed in to gcloud
[OK] a project is selected
[FIX] link billing: Billing > Account management
[FIX] create a budget: Billing > Budgets & alerts
ERROR: (gcloud.services.list) [you@example.com] does not have permission to access projects instance [my-project-typo-123] (or it may not exist): Project 'my-project-typo-123' not found or permission denied.
Help Token: AaVM3aPcsyqofgFCz8rvKX2Tc_3uhdkFVsqzYUr9NdJBvwpffVit9hWGZmVZIsFxvneJDdYgJ-TZmBKkIO9kY4sH1nwWYSL16EEL6TEmZysZsC7t. This command is authenticated as you@example.com which is the active account specified by the [core/account] propertygcloud is signed in, but every project check fails: Project 'my-project-typo-123' not found or permission denied. Nothing is broken; you are just talking to a project that does not exist (or that you cannot see).
Fix it: list your projects and use the exact id
# Fix it: list the projects you really have and use the right id (here: the lab project, shown as your-project-id)
gcloud projects list --filter="projectId=$PROJECT_ID" --format="table(projectId, lifecycleState)"
CLOUDSDK_CORE_PROJECT=$PROJECT_ID bash readiness.sh 2>&1 | head -3PROJECT_ID LIFECYCLE_STATE
your-project-id ACTIVE
[OK] signed in to gcloud
[OK] a project is selected
[OK] billing is linked (needed even for free tier)
gcloud projects list shows the real id (your-project-id here), and with it the checks pass again. Use the id column, never the name.
Part B: the 12 weeks
Each week: learn the idea (2 to 3 hours), do the lab (2 to 3 hours), write your note (30 minutes). Links go to our tested lab posts.
Block 1: foundations (Weeks 1 to 3)
| Week | Topic | Lab |
|---|---|---|
| 1 | Google Cloud basics: projects, billing, console, Cloud Shell | Google Cloud for Absolute Beginners: Projects, Billing and Console |
| 2 | Linux on a free VM | Linux for Cloud Beginners: 50 Commands on a Free VM |
| 3 | Bash and Git | Bash Scripting for DevOps and Git for DevOps |
Outcome: you can create and delete resources from the command line, work on a Linux server, script small tasks and keep your work in Git.
Block 2: core cloud services (Weeks 4 to 6)
| Week | Topic | Lab |
|---|---|---|
| 4 | Networking: VPC, subnets, firewalls | Google Cloud VPC for Beginners |
| 5 | Identity and security: IAM, secrets | Google Cloud IAM for Beginners and Secret Manager |
| 6 | Storage and data | Cloud Storage Deep Dive and BigQuery for Beginners |
Outcome: you can explain and build a private network, give least-privilege access, keep secrets out of code, and store and query data.
Block 3: delivery and operations (Weeks 7 to 9)
🎓⭐ Recommended by CareerByteCode · Live trainingAzure Cloud Architecture and AIOps Bootcampโน10,620See the training →| Week | Topic | Lab |
|---|---|---|
| 7 | Web servers and apps | Nginx for Beginners: Web Server, Reverse Proxy and HTTPS |
| 8 | Containers and CI/CD | Cloud Build Hands-On and CI/CD for Beginners |
| 9 | Logs and monitoring | Cloud Logging Hands-On and Monitor Linux Servers |
Outcome: you can deploy an app automatically from Git, find errors in logs, and set up monitoring and alerts.
Block 4: specialise and show your work (Weeks 10 to 12)
| Week | Topic | Lab |
|---|---|---|
| 10 | Kubernetes basics and GitOps | Argo CD on Kubernetes: GitOps from Zero and Prometheus and Grafana on Kubernetes |
| 11 | AI on Google Cloud | Gemini API for Beginners |
| 12 | Cost control and your portfolio | FinOps for Beginners: Read and Cut Your Cloud Bill, then turn your 12 notes into 3 project stories |
Outcome: you have run a Kubernetes workload, an AI call and a cost review, and you have notes for 12 labs to talk about.
What to do after Week 12
- Pick one direction and go deeper for 4 to 6 weeks: DevOps (more CI/CD, Kubernetes, Terraform), data (BigQuery, pipelines), or AI (Vertex AI, LLM apps).
- Certification: Google Cloud's Associate Cloud Engineer matches this roadmap well. Take the practice exam after Week 12; your lab notes are your revision list.
- Portfolio: pick your 3 best labs, write what you built, what broke and what you measured. Put them on GitHub or a blog. Interviewers ask "what did you build?", and now you have answers.
POC: prove it works
- The readiness script checked sign-in, project, billing, budget and APIs, then created and deleted a free VM and a bucket, leaving 0 resources.
- A project id typo made every project check fail; listing the projects and using the exact id fixed it.
- Every week of the roadmap links to a lab we tested on Google Cloud.
Quiz task: check what you learned
- What should you set up before your first lab? (Answer: billing and a budget alert.)
- The readiness check says
Project ... not found or permission denied. First thing to check? *(Answer: the project id withgcloud projects list.)* - What makes a lab "done" in this roadmap? (Answer: you can explain every step, it is cleaned up, and you wrote a short note.)
- Why write a note after each lab? (Answer: notes become your portfolio and interview stories.)
- Which certification fits this roadmap? (Answer: Google Cloud Associate Cloud Engineer.)
Clean up
The readiness script deletes what it creates and prints what is left (0 VMs and 0 test buckets above). Run gcloud compute instances list and gcloud storage buckets list at the end of every lab week to be sure.
FAQ
I have a full-time job or college. Is 6 to 8 hours a week enough? Yes. That is one evening for the idea and one for the lab, plus a short note. If a week is busy, repeat it the next week instead of skipping.
Do I need to know programming? Basic scripting helps (Week 3). Most cloud engineer work is configuration, the command line and reading code, not writing large programs.
Google Cloud, AWS or Azure? The ideas are the same everywhere (networks, identity, storage, compute, CI/CD). Learn one well first; the second is much faster.
Will I spend a lot? No, if you clean up after every lab and keep a budget alert. Most labs here run on Always Free or cost a few cents.
What if a lab fails for me? Good: read the error, compare with the Troubleshooting table in that post, and write down the fix. Fixing things is the job.
How do I show this to employers? Three short project write-ups with real numbers (what you built, what broke, how you fixed it), linked from your resume. Our post on talking about cloud projects shows how.
Troubleshooting
| You see | Why | Fix |
|---|---|---|
Project ... not found or permission denied | Wrong project id, or no access | gcloud projects list, then gcloud config set project <id> |
[FIX] link billing | No billing account on the project | Billing > Account management > Link a billing account |
[FIX] create a budget | No budget alert yet | Billing > Budgets & alerts > Create budget |
[FIX] compute API | The API is off (new projects) | gcloud services enable compute.googleapis.com |
| The VM test takes long or fails | API just enabled, or a quota | Wait a minute and run again; check IAM & Admin > Quotas |
What next
- Start Week 1: Google Cloud for Absolute Beginners: Projects, Billing and Console.
- More hands-on posts: ByteBlog. Practice in ByteLabs and see the courses.
Prepared by the CareerByteCode Team from our tested lab, with AI help for the write-up.
⭐ Keep learning with CareerByteCode
Hand-picked to go deeper on this topicGet CareerByteCode Team's next blog by email
One email a week when they publish. Add the weekly Best of ByteBlog too. No spam, unsubscribe with one click.
We email you once to confirm. By subscribing you agree to the privacy policy.
Related blogs
CareerByteCode TeamHow to Talk About Your Cloud Projects in Interviews
What you build: a STAR answer for cloud interviews from a real lab: a Cloud Run load test (0 failed, p95 64 ms), a broken release caught on its tag URL, and a script that uses only measured numbers.
#beginners
CareerByteCode TeamDevOps Engineer Roadmap 2027 with Hands-On Labs
What you build: a 2027 DevOps roadmap in 7 stages where every stage links to a tested lab, plus a real 15-minute lab that runs the whole loop: code, build, deploy, observe and roll back.
#ai
CareerByteCode TeamAI Engineer Roadmap: From Python to Production LLM Apps
What you build: a 7-stage AI engineer roadmap from Python to production LLM apps, with a real lab: a Gemini call, schema-checked JSON, a golden-set evaluation (6/6) and a temperature test.
#beginners
CareerByteCode TeamKubernetes Roadmap: From Pods to Platform Engineering
What you build: an 8-level Kubernetes roadmap from pods to platform engineering, with a real script that checks every level on GKE (pod, service, config, probes, HPA, RBAC, Helm, as code).
💬 Comments (0)
Top · NewestSign in to comment