What is our primary use case?
Our primary use case for SageMaker is for developing end to end machine learning solutions and ready solutions for things such as computer vision or speech recognition or speech to text. It's basically providing off-the-shelf solutions. Our customers are generally medium to enterprise size companies. We're a partner of Amazon.
What is most valuable?
The most valuable feature of the solution is that it allows you to create API endpoints and that saves a lot of time for data scientists.
What needs improvement?
The product has come a long way and they've added a lot of things, but in terms of improvement I would like to probably have features such as MLflow embedded into it.
Additional features I would like to see would include, as mentioned, MLflow and ML Pipelines which are more of a feature rich support of machine learning pipelines as well as scheduling machine learning pipelines, and visualization of machine learning pipelines.
For how long have I used the solution?
I've been using this solution for about a year.
What do I think about the stability of the solution?
The solution is quite stable.
What do I think about the scalability of the solution?
The solution is hosted on Amazon so it's quite scalable.
How are customer service and technical support?
The documentation is good so I haven't needed to use technical support.
Which solution did I use previously and why did I switch?
SageMaker was the first cloud solution I've used but there are other products, such as Databricks or Google and Azure that have similar products. There are common features with all these products but I'd say that SageMaker has more features than Databricks. Azure has other features in addition to Databricks, but SageMaker has provided everything.
How was the initial setup?
Initial setup is quite straightforward.
What's my experience with pricing, setup cost, and licensing?
The pricing for the Notebook endpoints is a bit high, but generally reasonable.
What other advice do I have?
I think for anyone using SageMaker it will help automate pipelines, and make it easier than doing the process manually. For anyone already on the AWS platform, they should definitely make use of it.
I would rate this product an eight out of 10.