Hard to onboard, great API product from Google!
Rating: 8 out of 10
IncentivizedUse Cases and Deployment Scope
We use the AutoML Natural Language part of the Google Cloud AI products. The NLP processor offered helps us understand user intent, sentiment, and perform other analysis on raw text received through multiple frontends for some of our experimental products. We also use the Cloud Vision API for street sign recognition, surveying, among other uses of converting text in images to text our backend can process and catalog.
Pros
- New products - Google is constantly releasing and adding new products to this API, it seems to be on of the fastest-growing products for this.
- Speed - The API is a lot faster than most of the alternative Computer Vision and general-use Machine Learning APIs out there.
- Comprehensive results - The results the API returns for most of the products don't require recurring API requests for processing. Everything is included and organized exactly in the JSON response as mentioned in their documentation for most of these Cloud AI products.
- The documentation format for this API is much better than some of Google's other documentation.
Cons
- Hard to find what to use - To find the right products, you need look closely at the details of each API, and find which suits your purposes. This can be easily fixed by creating a main page that details all of the products simply.
- Expensive - The API costs can quickly add up, especially during the setup process and as engineers figure out the usage of the API.
- No playground or training - There is a lack of an "API playground" or training sessions that could make onboarding engineers to this API much easier.
Likelihood to Recommend
Google Cloud AI is a wonderful product for companies that are looking to offset AI and ML processing power to cloud APIs, and specific Machine Learning use cases to APIs as well. For companies that are looking for very specific, customized ML capabilities that require lots of fine-tuning, it may be better to do this sort of processing through open-source libraries locally, to offset the costs that your company might incur through this API usage.
FP
Fedor Paretsky
Co-Founder/CEO in Engineering at aspace, Inc. (Computer Software, 1-10 employees)
Vetted Review
1 year of experience