The use of modern cloud services offers interesting opportunities in terms of cost, administrative ease and scalability. You only pay for the number of seconds your code is actually running, which makes it in general much cheaper than hosting your own server 24/7.
In order to capture most of what the serverless environment has to offer, all of the computing has to be done within a very resource-limited environment. A challenging constraint for most machine learning (ML) applications.
At Ixor we have successfully deployed some of our ML applications into the AWS cloud, in which all the training and inference is done in AWS Lambdas, the smallest computing component of the AWS offering. The type of algorithms that have been successfully deployed vary from random forests to medium sized neural networks for image classification and document named entity recognition.
If you are interested in the technical aspects of how to squeeze ML applications in AWS Lambda functions, you can read the full blog of our tips and tricks on our IxorThink Medium blog.
Feel free to contact us, if you are still puzzling with some unsolved questions on the matter.