Let’s say you have a Step Function running on AWS, and this workflow has CloudWatch alarms that notify you of successes, failures, and so forth. Generally speaking, CloudWatch alarms are set up with Amazon SNS to send email notifications. This is done by subscribing to an SNS topic and setting up the topic to email specified recipients when the CloudWatch alarm changes state (in this case, a Step Function success or failure is considered a change of state).
A few years ago I learned that you can extend the pandas library with your own methods, and this has been a real game-changer for the way I think about my pandas utilities. Essentially, a pandas “extension” is the process of adding your own namespace to pandas objects, such as Series and DataFrame objects. This allows you to access the methods on that namespace when you reference the object. One such example that comes baked into the library is the str namespace, accessible via pd.Series.str, which gives you access to a broad range of vectorized string functions on Series and Index objects.