Best Practices to Build Scalable AI Solutions on AWS
FAll 2023
During the live webinar by speaker Laskowsky, I was able to get an overview of what AWS offers that was not mentioned in class, especially the Machine Learning operations and Data pipeline aspect of AWS. During data preparation, it is critical to understand concepts such as data storage and accessibility, data cleaning, sampling, schema definition, and data versioning. Data needs to be organized in order to find how the data would be helpful in different concepts. Then the idea relates to the ML pipeline that typically consists of several stages to move data from raw data to a trained model that can then be used to make predictions or classifications. Some key components that stood out to me were data collection and ingestion, model training, evaluating, deployment, and the security and compliance aspects.