Don’t dip your toes in serverless — you have to dive right in
Paul Johnston

Can you tell me more about the impact of a classic software project? Usually we have a software lifecycle of build, test (different stages), deploy (blue / green) maybe paired with canary releases.

Software evolves and needs to change over time. Therefore a fast test execution for adaptions and refactorings are mandatory. For non serverless implementations using known best practices this works very well in a lot of different languages.

Going for serverless for now sounds like a decision for production only. Therefore instead of building an artifact to deploy and handle the deployment explicitly, serverless adds a bit more complexity.

Nowadays you put your piece of software in an isolated unit. You can run it on dev, test and prod systems quiet similar. Now, how comes serverless into play. As I know there are a few ways to simulate AWS / Azure like environments on non cloud environments. But for now I never read a good report about how professional and efficient software development works in combination with serverless elements. Especially for complex projects this seems to be an unsolved challenge.