Is monitoring separate from testing? Should it be?

I realized I thought of e2e tests as always being pretty intimately connected with both monitoring and testing. Especially if you take the time to write your tests in something like Playwright, and they're doing complex user behaviors and measuring response times, you want those running both as part of QA/Test and as part of your general monitoring.

I've recently had more than one discussion where it seemed like the distinction was very clear, and that stuff like Canary deploys were across the line from QA and now part of monitoring for the SRE/Operations.

I have two questions:

* is there a reason that things like Synthetic user testing shouldn't be used at the QA/Test/Staging, hell even Dev deployment stages?

* Should developers be writing the tests that will later be used to do uptime and synthetics checks? Or is that monitoring better left to the SREs and Ops specialists?

On the other hand is there a reason on the operations side that you wouldn't want to run the detailed tests that were written for QA?

Full disclosure I work for Synthetics tool Checkly, not trying to promote Checkly just hoping to understand this a bit better.