Surviving Scientific Review

The two reference tools at the back of the book can now be used on a screen at intellectualfrontiers.com. The PI Award-Intelligence Checklist is a task list of 37 questions in seven groups, from deciding whether to write through what to learn after the review, and you can copy it into your project notes and tick it off against an application already in motion. The Labor as Code Skill Catalog is a table of the 21 skills an Award Intelligence AI Workforce would need, from the Scientific Proposition Extractor to the Outcome Reconciler, with what each one does. Both are also still printed in the book’s appendices, word for word.

NIH received 62,592 Research Project Grant applications in fiscal year 2025 and funded about 13 percent. Peer review is careful and rule-bound, but a scientist meets it from the outside, a few applications at a time, while NIH and experienced institutions work from knowledge most PIs never see. This book treats that gap as a problem you can solve without gaming the process. It shows how to scout the review environment before writing a page: understand NIH's rules, study funded analogs and use public data and AI. You learn to choose the mechanism, Institute and study section your science fits, turn preliminary evidence into a proposal reviewers can evaluate, and run a red team and mock review before NIH sees the application. It covers Program Officer conversations, reading a summary statement for what it says, and building the institutional memory that makes every next application stronger. Eleven chapters.