This is pretty much the exact target audience of the book that I'm
working on very slowly.
For now, I'd suggest starting with:
1. See
"http://docs.racket-lang.org/getting-started/index.html".
2. Do little one-hour ``pilot projects'' and experiments.
3. Ask questions on this email list.
A few months is a luxurious lead time to prep for a one-person project
that will be on a schedule. Or, if you'll have a project team, a few
months gives you some time to start to become a guru on the
technology. However, if you have a team project, and you have some
tricky problems to solve, or you want to get a team up to speed more
quickly and reduce the number of wrong turns, I think one good way is
to have a consultant help. (Full disclosure: I make a living being
such a consultant.) You could also rapidly own expertise by hiring a
new employee: a seasoned expert, or a promising new grad who's been
using Racket.
Alternate strategy: if you want to be coding on your project in Racket
by lunchtime, just pretend it's Pascal with a Lisp syntax, and start
typing. You probably won't get the big wins that way, but I've seen it
done successfully wrt business goals.
Neil V.