A little over a year and a half ago, we started diving deep into the serverless space. We spoke to many users of AWS Lambda and found that the biggest issue with building complex serverless application was a lack of visibility and instrumentation.
It's difficult to know when issues start bubbling up, and relying on so many services as part of your application can be scary. We've built a set of tools to help dig deep into functions, and view outwards by tracing triggers and external calls performance and metadata surrounding it.
I'd love to hear more from Lambda users about their experiences building and operating serverless apps.