If an error is thrown in a deployed serverless application, but no agent records it, did it cause customer frustration? If your app is not observed, does it exist in a quantum state of working and breaking simultaneously?

We've all been here before: apps that have been deployed but not optimized not only impact performance but can also cost you resources, customer goodwill, and ultimately money. Join us on a journey through the serverless cosmos, looking at ways to observe, debug, and optimize a serverless application (in Python).