[2OEH8eoCRo0]

This was my work in defense. Weeks or months are spent in design and the code itself only takes a few days.

Now at my company teams will spend no time planning, code like crazy, only to redo it multiple times.

Illusion of progress. There must be a middle ground.

[necovek]

As mentioned below, it seems you don't like true agility (though, the critical piece is re-evaluating often, most teams miss that).

If you want a successful example in a traditional engineering industry, you don't need to look further than SpaceX vs Boeing and their rocket development (one got a smaller budget and blew up a number of rockets and has been earning money for 4+ years, the other got twice the grant and wasn't trusted to bring astronauts back from ISS a few months ago).

[azemetre]

Is this the core tenant of agile development? To release often and speak to stakeholders often? The hardest part is deciding when to do that first release I suppose.

[necovek]

As soon as you've got something releasable.

Though, you should break work up in a way to get to something releasable asap.

The latter is where I think it's more art than science still, or at least I can't come up with a good written process on how you do it (other than the constant "where is the value in this" and "what's the smallest thing we can build" questioning), but I can always do it!