Resources for the Google Cloud Professional Machine Learning Engineer Certification

Recently I passed the Google Cloud Professional ML Engineer Certification Exam and received a few queries on how to prepare for the exam. Hence I am writing up a post to share some free/low-cost training offers and resources that have helped me pass the exam.

One thing to note is that I have previously completed the Google Cloud Professional Data Engineer Certification and have some prior experience with using Google Cloud Platform and TensorFlow/Keras before attempting this certification exam. I shall suggest some resources for learning the portions related to Data Engineering and TensorFlow if you are not familiar with them.

There are already some folks who have completed the certification exam and have kindly shared their experiences. These are the articles that I reference:

Article References

• Google Cloud Professional Machine Learning Engineer Certification Preparation Guide by Dmitri Lerko and Steven Macmanus
• Cloud Professional ML Engineer by Viren Radhakrishnan
• How to Ace the Google Cloud Professional Machine Learning Engineer Certification by Salem Amazit

You should definitely check out the official guides and sample questions linked below.

Official References

• Official Google Cloud Professional ML Engineer Certification Guide
• Sample questions
  • I recommend to attempt sample questions early to identify concepts that need brushing up and have an expectation of the style of the questions asked.

Below are some free/low-cost training offers and books that I use for preparation or revision.

Courses/books that I used for preparation/revision

• Google Machine Learning Crash Course and Guides
• Training offer from Google Cloud for PluralSight, Coursera and Qwiklabs
  • Between Coursera and PluralSight, I would suggest to take up the PluralSight offer as it offers access to the entire Google Cloud Catalog for a monthly subscription at 19 USD/month rather than paying by each Coursera specialization
  • Direct PluralSight Google Cloud Catalog Sign-up link
  • Note that some courses such as Advanced Machine Learning have not been updated fully to TF2 so the code explanation sections can be skipped.
  • I revised/took the PluralSight courses below
    • Machine Learning on Google Cloud
    • Advanced Machine Learning on Google Cloud
    • [NOTE] There are new courses that were released posted after I took my exams but they seem relevant
      • MLOps (Machine Learning Operations) Fundamentals
      • Managing Machine Learning Projects with Google Cloud
  • Qwiklabs
    • The offer also includes a free 1-month subscription that allows you to complete quests and have hands-on practice
    • Should this offer lapse, I would suggest you to keep a look out for future training offers like majulahgcp (Current season is already over) for free qwiklab subscriptions
    • I shall mention the applicable qwiklabs you can take to get hands-on practice for specific sections in the exam guide
• The book "Building Machine Learning Pipelines" (BMLP) by Hannes Hapke and Catherine Nelson is also a great reference
  • You can consider borrowing the e-book for free from National Library Board Singapore if you are a member.
  • While the official tutorials (linked in relevant sections below) allows you to do a hands on practice with the TFX components like TF Data Validation and TF Model Analysis, I feel that this book provide a better and more concise overview of how they work and how to use each component
  • They also describe in great detail how the other pipelines in Kubeflow and Airflow can be set up

Additional resources

Below are some additional resources that I focused for the respective sections in the exam guide.

Please note that this should be read in tandem with Dmitri Lerko and Steven Macmanus’ guide mentioned above for a more comprehensive coverage/revision depending on your experience level/background.

Problem Framing

• Introduction to ML Problem Framing
  • This provides an approach for taking requirements and analyzing them to determine if AI/ML provides value to the business

ML Solution Architecture

• Overview of Google Cloud Products
  • AI/ML Products
  • Data Analytics Products
• Google Cloud ML Solutions
  • Architecture for MLOps using TFX, Kubeflow Pipelines, and Cloud Build
  • Best practices for performance and cost optimization for machine learning
  • Building production-ready data pipelines using Dataflow: Overview
  • Minimizing real-time prediction serving latency in machine learning
  • Rules of ML

Data Preparation and Processing

• Data Preparation and Feature Engineering in ML
• If you want to deep dive, you can look at the Pluralsight Data Engineering on Google Cloud Platform course
• Qwiklabs Quests
  • Scientific Data Processing
  • Data Engineering Quest (only if you want to deep dive)

ML Model Development

• Understand use cases for Custom Model Development with TensorFlow and using products like AutoML, AutoML Tables, BigQuery ML or Google Cloud AI APIs
• Debugging and Testing section in Testing and Debugging Guide
• If you want to deep dive into Model Training with TensorFlow 2 and tf.keras, either use the Plural sight courses mentioned above or use these free resources
  • Machine learning Crash Course mentioned above
  • Keras.io Guides
  • TensorFlow Official Guides
• Qwiklabs Quests
  • Integrate with Machine Learning APIs
  • Intermediate ML: TensorFlow on GCP
  • BigQuery for Machine Learning

ML Pipeline Automation & Orchestration

• Introduction to Google Cloud AI Platform
• Video series on Kubeflow and Kubeflow Pipelines: KubeFlow 101 Playlist by Stephanie Wong
• Understand the overall solution with Chapters 1 and 2 in the BMLP e-book
• Run the following notebooks in Google Colab!
  • TFX AirFlow Workshop
  • TFX for TF2 Keras Tutorial
  • TFX for TF Estimator Tutorial
  • TFX on Cloud Platform Pipelines
• Deep Dive into relevant chapters in the BMLP e-book as required
• Qwiklabs Quests
  • Advanced ML: ML Infrastructure
  • KubeFlow Pipelines with AI Platform
• ML Solution Monitoring, Optimization, and Maintenance ML Engineering
• Testing in Production section in Testing and Debugging Guide

System Biases, Fairness, Interpretability

• Integrated Gradients TensorFlow Tutorial
• Introducing the What-If Tool Playlist by Sara Robinson and James Wexler