What it is
The Professional Machine Learning Engineer certification is Google Cloud's own credential for engineers who "build, evaluate, productionize, and optimize AI solutions using Google Cloud capabilities" (Google Cloud's certification page). It's an assessment, not a taught course: candidates study independently, then sit one proctored exam.
Format and syllabus
The exam has 50-60 multiple-choice and multiple-select questions in a 2-hour sitting, offered in English or Japanese. Google's certification page groups the content into six domains: architecting low-code AI solutions, collaborating to manage data and models, scaling prototypes into ML models, serving and scaling models, automating and orchestrating ML pipelines, and monitoring AI solutions. Google states the exam has been updated to reflect the shift from Vertex AI toward its Gemini Enterprise Agent Platform, plus changes to its data and analytics stack. The full topic breakdown is in Google's published exam guide PDF, linked from the certification page.
Prerequisites and who it suits
There's no formal entry requirement, but Google recommends 3+ years of industry experience, including at least a year designing and managing ML solutions on Google Cloud - so it targets practising ML engineers rather than beginners.
Booking and price
Registration costs $200 plus tax where applicable, booked through Google's CertMetrics portal. Candidates sit it either online-proctored via Pearson OnVUE from any location meeting the testing requirements, or onsite at a Pearson VUE test centre; Pearson VUE's UK site (pearsonvue.com/gb) lists centres and a locator, so it can be booked and sat from the UK. The certification is valid for 3 years, after which it lapses unless renewed by exam.
