The one thing about large enterprise projects is though, you can usually convince the mismanagers that it's good for their career, and the mismanagers end up disliking the competent programmers. If you're a vulture or a hyena a large IT project can become like a elephant corpse, easy pickings. Convince the mismanagers to move competent people to your team in exchange for your yes men. Then find a piece of infrastructure you need for your project and offer to build it for the larger project. This is basically an excuse to use their infrastructure, db servers, test servers, etc. Use this to get buy in to slip your own schedule so you have time to deal with the inevitable feature creep. Go to the design and planning meetings and suggest new and exciting technologies that they can figure out. When it doesn't work, take the hardware they req'd off their hands for the infrastructure you offered to build them. Make sure you integrate your project to the infrastructure first (you can always pitch it as reducing risk to the large project) so they can't make any changes to it with out breaking production code.
But why would I want to do all these things for the benefit of a toxic organization? That system feeds off people who try to do the right thing, then it chews them up and spits them out. In fact it relies on there being people who will do the right thing even tho' it means extra stress and no extra reward. Organizations like that deserve to collapse under their own weight.
... For the benefit of the management tier above that allowed the toxic situation to develop in the first place. You can bet that you won't see any of the profits yourself, and they're laughing all the way to the bank, they took their eyes off the ball and still scored.
