For people running AWS Lambda in production:

How do you currently handle Lambda errors before users complain?

Do you just use CloudWatch alarms/logs, or do you have something cleaner?

I’m asking because digging through CloudWatch log streams, grouping repeated errors, and sending useful Slack alerts feels more painful than it should be.

Curious what your setup looks like:

- CloudWatch only?

- Datadog / Dashbird / Lumigo?

- Custom Lambda + SNS + Slack?

- Something else?

What’s the most annoying part of your current setup?