We'll be holding a 'Project Applications' sprint during DrupalCon Denver, and I wanted to start a topic to discuss potential items of focus.
To begin, I see two main thrusts for a project applications sprint. The first goal would be to solicit additional reviewers and attempt to make a significant dent in the 'Needs Review' queue. However, as this is just a temporary fix (as we've seen with the queue once again growing ot over 150 'Needs Review' applications), there is a second potential goal of sitting down and coding the automation and process changes which are needed to transition the Project Application process to a state where it is manageable over the long term.
For me, I see some potential tasks which could fall into this second category:
- Enhancing the testing infrastructure to support testing of 'sandbox' projects
- This has heavy project_issue dependencies, such as the fact that sandbox issue queues do not have a assigned software version, so there's no way to tell the testing infrastructure what repository branch/tag it should be checking out for test.
- Develop a new 'project-apps' test type
- This would be a new test type (to supplement the current testing infrastructure's 'file' and 'branch' test types), which would need to be overlaid/integrated into the existing PIFT module.
- Develop a new 'project-apps' PIFR client plugin
- This would essentially be an instance of PAReview and/or DrupalCS, but integrated with the existing testing process (and thus hosted on official d.o infrastructure).
- Further develop the drupalorg_projectapps module
- This would be a central drupal.org container for all projectapp-specific customizations. I've got a start within a sandbox at http://drupal.org/sandbox/jthorson/1367220 ... but, being slightly more aggressive, this could be a module container for a whole new front end user experience for new project application participants, guiding them through and ensuring they've covered the bases before allowing them to even create the application ticket. (See the mockups in the sandbox issue queue for an example of what this could look like.)
- Develop a 'Project Dashboard' to appear on project sandboxes (or all projects)
- This dashboard could provide information on testing results, repository health, and other automated checks which are often validated through the project application process. As this is useful information for all projects, it should potentially be coordinated with the Prairie initiative.
- Work on improving DrupalCS/PAReview projects
- As tools which have revolutionized the review process, we could also consider the enhancement of these tools as well as a potential sprint focus.
EDIT (added after Klausi's comment below):
Lets start with a brainstorming session ... what else might we want to try and accomplish during a ProjectApps sprint? What else could we accomplish with a drupalorg_projectapps module? What would be the 'ideal' on-ramp user experience for new applicants? What can we do to make the reviewer's jobs easier?
For now, be bold, be creative, and be outrageous ... we'll narrow the list down to some primary objectives and tasks after a few days of brainstorming.

Comments
I'm not sure yet if I will
I'm not sure yet if I will attend this code sprint or another one, so I don't want to make any promises. But if I join you I would like to introduce others to manual project application reviewing and attack the project application issue queue a bit. I don't have much interest in working on PIFR or project.module components directly, but I could help with writing integration for pareview.sh or drupalcs. And of course I would like to work/brainstorm on drupalcs and how we could even detect more common Drupal coding mistakes with automated tools.