Home/Learning paths/Google Cloud

GOOGLE CLOUD · PROFESSIONAL

Professional Machine
Learning Engineer.

Understand the whole ML lifecycle. Build the judgment to choose the right solution. Start with focused guides, then deepen your practice with official labs.

Start with the fundamentals ↗ Browse study guides ↗

EXAM DURATION2 hours

QUESTION FORMAT50–60 questions

REGISTRATION$200 + applicable tax

STUDY RHYTHM8 weeks, at your pace

Exam facts checked November 10, 2026 on the official certification page. The current guide uses Gemini Enterprise Agent Platform terminology; older resources may refer to Vertex AI. Our eight-week plan is a suggested routine, not an official requirement or a guarantee of readiness.

YOUR LEARNING ROADMAP

Your guides. One connected picture.

CONSISTENCY OVER CRAMMING

Your eight-week study plan.

Use each week as a focus, not a deadline. Read the guide, complete its exercise, then extend your learning with official documentation and hands-on labs. Your checklist is saved in this browser.

0 of 8 weeks completed

Week 1Frame ML problems and choose a baseline.
Read guide 01. Define a business metric and compare low-code, custom, and foundation model approaches.Week 2Build reliable data and feature workflows.
Read guide 02. Design a split, test for leakage, and document an experiment.Week 3Train, evaluate, and tune.
Read guide 03. Compare a baseline with a tuned candidate using a fixed evaluation set.Week 4Design a serving strategy.
Read guide 04. Create a batch or online serving plan and define a rollback condition.Week 5Connect the ML lifecycle.
Read guide 05. Sketch a pipeline with validation, evaluation, and promotion gates.Week 6Monitor quality and manage risk.
Read guide 06. Design a monitoring dashboard and a generative AI evaluation rubric.Week 7Practice across the full blueprint.
Work through official labs and sample questions. Map weak areas to the current exam guide and revisit those topics.Week 8Review your decisions and close the gaps.
Explain end-to-end scenarios aloud. Recheck current exam details and review your notes before deciding whether to book.

CONTINUE LEARNING

Put your knowledge to work.

Explore labs, original scenarios, and quick references.

Browse study guides ↗

Professional MachineLearning Engineer.

Your guides. One connected picture.

Choosing your ML approach

Data, features & experiments

Training & model development

Serving & scaling models

Pipelines & MLOps

Monitoring & responsible AI

Classification metrics & thresholds

Drift, skew & monitoring

Point-in-time features

Grounded generation & RAG

Safe model rollouts

Compute & hyperparameter tuning

Pipeline gates & lineage

Fairness & explainability

Low-code AI service choices

ML workflow security

Prompting, retrieval & tuning

Forecasting & backtesting

Batch & streaming features

Generative AI evaluation

Rare events & calibration

Registry, experiments & lineage

Inference latency & cost

Feature Store fundamentals

Diagnosing model errors

Labels, sampling & bias

Embeddings & vector search

Online model experiments

Distributed training

Custom serving containers

Regression metrics & residuals

Clustering & segmentation

Preprocessing & parity

Evaluation uncertainty

Anomaly detection & alerts

ML workflow orchestration

Sensitive data in ML

Tabular model selection

Image data & evaluation

Feature attribution

Transfer learning

Recommendation systems

Data contracts

Weighting & resampling

Model quantization

Ranking metrics

Data augmentation

Pipeline cache correctness

Your eight-week study plan.

Put your knowledge to work.

Professional Machine
Learning Engineer.