The Importance of Observability in ML Product Development

I like to think about a port as the mix panel for AI. So our users, where they're data scientists, ML engineers or product managers, can see what models are running currently in production and how well they are performing. And then it's the PM's role to take that and compare it to the business stack. You should be able to measure them as your model is running in production. This is kind of where a port you fit into the picture. The same thing happened with observability and ML product. It's almost most pure form of it.