Note (2026-05-06): Vertex AI was rebranded as Gemini Enterprise Agent Platform. Google's exam guides still use the Vertex AI naming, so this article does too. The official guides may switch to the new name at some point as you prep, but for now we're matching the language currently in the exam materials.
A lot of people come to me asking what's on the new Google Cloud Professional Data Engineer exam.
Google has changed this exam significantly recently.
They did this to better align the exam with the skills and technologies that data engineers are expected to use in modern cloud environments.
This article provides a logical overview of what has recently changed, what topics were reduced or removed, and what areas are now emphasized.
Beware of outdated courses out there, and use this article as a guide for what's relevant.
Several services and topics that were heavily featured on the old exam have been significantly reduced or removed. Key examples include:
Overall, the exam now places less emphasis on infrastructure management and machine learning engineering.
The updated exam focuses more on real-world, business-driven data engineering solutions. Candidates are expected to understand how to architect, operationalize, and govern modern data systems. Key areas of increased focus include:
Understanding how to securely and efficiently share data across teams, organizations, and clouds is now a core skill. Key services include:
These services reflect the industry's move toward distributed, multi-cloud data ecosystems.
Building data pipelines with low-code solutions is increasingly important. Candidates should be familiar with:
These tools enable faster development cycles and integration without extensive custom coding.
Beyond building solutions, candidates must understand how to operationalize them in production environments. Topics include:
The focus is on ensuring that data pipelines and applications are reliable, maintainable, and scalable.
Candidates must demonstrate an understanding of fundamental cloud networking and security concepts, including:
Security is treated as a first-class concern, integrated into architecture rather than added later.
Proper governance of data resources is emphasized more heavily. Key services include:
Understanding these services is critical for designing scalable and compliant data architectures.
High availability and resilience design are major priorities. Candidates must understand:
These considerations are central to building fault-tolerant, production-ready systems.
One major change in the updated exam is a move toward greater breadth with less technical depth in individual services.
Technical knowledge remains important, but the exam now tests a candidate's ability to design complete solutions across multiple services rather than asking in-depth configuration questions about any single service.
For example, whereas the old exam might have required detailed knowledge of Kubernetes cluster settings or machine learning hyperparameter tuning, the new exam focuses more on how you would integrate services like BigLake, Dataform, and Dataplex into a cohesive solution to meet specific business requirements.
This change reflects the way that modern data engineering is practiced: building architectures that span services, manage risk, ensure resilience, and maintain governance, rather than deep specialization in one tool.
The expectation is that candidates can think architecturally and understand trade-offs across different GCP services.
The recent overhaul of the Professional Data Engineer exam marks a fundamental shift in how Google Cloud defines the role of a data engineer.
Candidates who are preparing for the exam today should adapt their study plans to focus on architecture patterns, governance, security, operationalization, and modern data sharing and integration strategies.
Detailed infrastructure management and machine learning workflows, while still valuable skills, are no longer the centerpiece of the exam.
Focusing on the new topics and services introduced in the exam guide, and practicing the design of real-world solutions that meet business needs, will position candidates for success.
Understanding these changes is critical not just for passing the exam, but for being effective in data engineering roles that increasingly demand broad architectural thinking, cross-team collaboration, and integrated cloud-native solutions.
By approaching your preparation with this mindset, you will be better aligned with the real-world expectations of modern data engineering roles in the cloud era.
