How Difficult is the Google Cloud Data Engineering Exam?

My personal journey + 10 tips to prepare for the exam

6 min readMar 28, 2022

Spanish version of this post here

It has been more than 2 years since I took the exam at the request of the company where I worked. Back then I only had a month to prepare. A lot? Too little? What do you think was the result?

Google recommends 3+ years of industry experience, including 1+ year building and managing solutions with GCP to be successful with certification. The truth is that I did not pass. And I must confess that this experience marked me so much that the following months I decided to study each and every specialization that was available at that time on Coursera. I studied and studied.

And I’m still studying…

Although sometimes things get in the way like conferences to which I am invited as a speaker, having to learn about DevOps, having to learn a new programming language, creating the largest and coolest data community in Latin America, and other things in life.

There must be a better way to optimize my time. According to my research, here are some tips that I have found on the web.

Update: In December 2022 I took the exam again and this time I passed!!!

Tip 1: You can’t study everything. Better study what they are going to evaluate you

According to Google Cloud, the definition of what a Data Engineer does is as follows:

The Professional Data Engineer enables data-driven decision making by collecting, transforming, and publishing data. The Data Engineer must be able to design, compile, operate, secure, and monitor data processing systems, with particular emphasis on security, compliance, scalability, efficiency, reliability, fidelity, flexibility, and portability. Additionally, the Data Engineer must be able to constantly leverage, deploy, and train pre-existing machine learning models.

Having made that clear, the following topics are included in the exam (divided into 4 sections):

Section 1. Designing data processing systems

Mapping storage systems to business requirements
Data modeling
Trade-offs involving latency, throughput, transactions
Distributed systems
Schema design
Data publishing and visualization (e.g., BigQuery)
Batch and streaming data (e.g., Dataflow, Dataproc, Apache Beam, Apache Spark and Hadoop ecosystem, Pub/Sub, Apache Kafka)
Online (interactive) vs. batch predictions
Job automation and orchestration (e.g., Cloud Composer)
Choice of infrastructure
System availability and fault tolerance
Use of distributed systems
Capacity planning
Hybrid cloud and edge computing
Architecture options (e.g., message brokers, message queues, middleware, service-oriented architecture, serverless functions)
At least once, in-order, and exactly once, etc., event processing
Awareness of current state and how to migrate a design to a future state
Migrating from on-premises to cloud (Data Transfer Service, Transfer Appliance, Cloud Networking)
Validating a migration

A lot? This is just section 1!

Section 2. Building and operationalizing data processing systems

Effective use of managed services (Cloud Bigtable, Cloud Spanner, Cloud SQL, BigQuery, Cloud Storage, Datastore, Memorystore)
Storage costs and performance
Life cycle management of data
Data cleansing
Batch and streaming
Transformation
Data acquisition and import
Integrating with new data sources
Provisioning resources
Monitoring pipelines
Adjusting pipelines
Testing and quality control

Section 3. Operationalizing machine learning models

ML APIs (e.g., Vision API, Speech API)
Customizing ML APIs (e.g., AutoML Vision, Auto ML text)
Conversational experiences (e.g., Dialogflow)
Ingesting appropriate data
Retraining of machine learning models (AI Platform Prediction and Training, BigQuery ML, Kubeflow, Spark ML)
Continuous evaluation
Distributed vs. single machine
Use of edge compute
Hardware accelerators (e.g., GPU, TPU)
Machine learning terminology (e.g., features, labels, models, regression, classification, recommendation, supervised and unsupervised learning, evaluation metrics)
Impact of dependencies of machine learning models
Common sources of error (e.g., assumptions about data)

Section4. Ensuring solution quality

Identity and access management (e.g., Cloud IAM)
Data security (encryption, key management)
Ensuring privacy (e.g., Data Loss Prevention API)
Legal compliance (e.g., Health Insurance Portability and Accountability Act (HIPAA), Children’s Online Privacy Protection Act (COPPA), FedRAMP, General Data Protection Regulation (GDPR))
Building and running test suites
Pipeline monitoring (e.g., Cloud Monitoring)
Assessing, troubleshooting, and improving data representations and data processing infrastructure
Resizing and autoscaling resources
Performing data preparation and quality control (e.g., Dataprep)
Verification and monitoring
Planning, executing, and stress testing data recovery (fault tolerance, rerunning failed jobs, performing retrospective re-analysis)
Choosing between ACID, idempotent, eventually consistent requirements
Mapping to current and future business requirements
Designing for data and application portability (e.g., multicloud, data residency requirements)
Data staging, cataloging, and discovery

More detail here.

Tip 2: Take notes

Don’t limit yourself to just reading or watching videos. Try to apply more active learning by taking notes in a notebook or using tools like Remnote

The first time you will see that you will spend a lot of time taking notes but the knowledge will be fixed more time than if you are only in listening mode. Remember, it’s not about watching more videos in the least amount of time. There will be an opportunity to watch the videos at 2x speed of the player later. In fact, this is best done when you already know the subject and are just looking to do a quick review.

Tip 3: Study for free here!

Google wants you to get certified so it offers you free training. I don’t know for how long this will be available, so go ahead and take advantage of it:

Study for free

Tip 4: Read books

This book is required reading:

Official Google Cloud Certified Professional Data Engineer Study Guide by Dan Sullivan.

Tip 5: Read about Best Practices

Tip 6: Get down to business with Qwiklabs badges (now called cloudskillsboost)

Qwiklabs is a learning platform that combines theory and practice through labs where you are given access to the GCP cloud without using your personal account. A series of labs make up a quest. This particular quest is what you need to prepare for the exam:

Engineer Data in Google Cloud

Sometimes there are events where credits are given away so you have to stay tuned.

Tip 7: Take the preparation program on Coursera

This program consists of the following 6 courses:

Google Cloud Big Data and Machine Learning Fundamentals
Modernizing Data Lakes and Data Warehouses with Google Cloud
Building Batch Data Pipelines on GCP
Building Resilient Streaming Analytics Systems on Google Cloud
Smart Analytics, Machine Learning, and AI on GCP
Preparing for the Google Cloud Professional Data Engineer Exam

Preparing for Google Cloud Certification: Cloud Data Engineer Professional Certificate in Coursera.

Tip 8: Get familiar with the format of the exam questions and the content that may come to you

According to each case, you will have to select one option. You get feedback on why and why not of each option when you finish answering all the questions. Here comes the trick. You should write down those topics where you did not do well (for example, Kubernetes) and study them more intensively.

Professional Data Engineer Sample Questions

Tip 9: Check out the resources available in the Learning and Certifications Hub

Learning & Certification Hub

Tip 10: Schedule that exam!

Finally, I once heard that a good technique to study conscientiously is to schedule the exam. Give yourself a deadline. That way you will have the pressure that you have to take that exam and you will not let time pass.

Do you have any other tip? Tell me in the comments.

For more information, I recommend the following:

More interesting courses listed here

Thanks for reading! Do you want more?

Hit 50 times the like button and something wonderful will happen.

👉Follow me for more nerdy talks!
👉Follow Data Engineering Latam for more content related to Data Engineering, Data Science and Data Management.