A week-long hackathon, sounds fun!! I think the most important thing to do for a hackathon is often to set and agree with the team on what will be presented at the end. I emphasize presented and not built, because the biggest impact you can have is to influence future activity, not to solve all problems in a week. The story you tell with your presentation will be determine the future of the project, and the week's work should be prioritized based on what will best support the story you want to tell.

That said, it can also be really fun to set the goal of actually solving something with the time you now have, and showing it off. Just try to avoid a compromise where you build something 50% to production and tell 50% of a compelling story... you'll end up with leftover, low-priority homework that doesn't fit into anyone's schedule