G2

Linda A.

Architect

Architecture & Planning

Small-Business (50 or fewer emp.)

9/1/2026

"Useful platform for building custom ML models with less coding overhead"

4.5/5

Describe the project or task Google Cloud AutoML helped with:

Service task overview not provided. Review collected by and hosted on G2.com.

What do you like best about Google Cloud AutoML?

As a solutions architect, I like that Google Cloud AutoML lowers the barrier for creating custom machine learning models without requiring a full data-science team for every use case. We’ve used it mainly for vision and tabular models training on domain-specific datasets and deploying them into existing cloud architectures. The managed training process, integration with Google Cloud storage and serving options, and relatively straightforward evaluation metrics make it practical when we need to move from prototype to production faster. From an architecture standpoint, it fits well when we want to keep models inside the Google Cloud ecosystem and control data residency and access. Review collected by and hosted on G2.com.

What do you dislike about Google Cloud AutoML?

It can become expensive at scale, especially with larger datasets or frequent retraining. Customization options are more limited compared to building models directly in Vertex AI or with open-source frameworks, so for complex or highly optimized use cases we still need to step outside AutoML. The learning curve around data preparation, labeling quality, and interpreting model metrics is higher than it first appears. Documentation is generally good, but advanced deployment and monitoring patterns sometimes require extra research. Review collected by and hosted on G2.com.

Recommendations to others considering Google Cloud AutoML:

Recommendations not provided. Review collected by and hosted on G2.com.

What problems is Google Cloud AutoML solving and how is that benefiting you?

It helps us deliver custom ML capabilities faster when full custom model development would take too long. For projects that need domain-specific vision or structured-data models, AutoML reduces the time from data collection to a working endpoint. This has allowed architecture teams to prototype and validate ML features earlier, then hand off to engineering for integration, while keeping the solution within a managed Google Cloud environment. Review collected by and hosted on G2.com.