There are a few big assumptions in here, that in less-than-optimal companies require time-eating validation: * The source you have corresponds to whatever runs

There are a few big assumptions in here, that in less-than-optimal companies require time-eating validation: * The source you have corresponds to whatever runs on production. No, you don't get access. * The source code has a clear, unique home ('here is a stack of DVDs with the full content of the last person's hard disk. No idea if someone else touched it and dropped it in some repo') * The source code compiles and runs locally. And doesn't damage production when you start it up because some monkey left a password in there. And doesn't shit all over your hard drive as said monkey expected a very specific folder structure on his personal laptop. * There is basic test infrastructure. Adding a very first test to an untested code base can take forever My experience is that most of this is OK when the software was developed internally. Internal people keep the nest clean enough, as they know they'll have to deal with the long-term consequences. When some random contractor or third party did a fly-by code drop('IT was too slow so business department X decided to hire someone themselves'), fear for your sanity.