Formulating NIH Budgets for Computer Science & AI

1. Financial Alignment & Eligibility Standards

Securing research funding from National Institutes of Health requires meticulous adherence to both financial eligibility standards and administrative regulations. For projects in the domain of Computer Science & AI, budgets must be constructed using realistic cost projections that are directly tied to the scientific methodology. Under-budgeting may jeopardise project execution, while over-budgeting or including ineligible costs often leads to immediate rejection during administrative screening.

Computational research in Computer Science & AI is heavily weighted toward high-performance computing (HPC) nodes, scalable cloud storage, specialized developer software, and travel for rapid presentation dissemination at international proceedings, which must be clearly justified to NIH reviewers.

Proposal teams must submit all budget items in the host institution's local currency, mapping them to the specific electronic submission environment (eRA Commons). Every cost item must be justifiable as necessary, reasonable, and allocable to the project.

2. Direct vs. Indirect Cost Categorisation

A primary point of auditing compliance is the strict division between Direct Costs (expenses directly attributable to the execution of the research project) and Indirect Costs (institutional overheads, facility maintenance, and central administrative support).

Institutional overheads (Facilities & Administrative - F&A) under **NIH** guidelines are calculated using your university's Negotiated F&A Rate applied directly to the Modified Total Direct Costs (MTDC). Under these regulations, major equipment over $5,000, sub-award amounts over $25,000, and student tuition must be excluded from this indirect cost base for **Computer Science & AI** awards.

For NIH proposals, the indirect cost rate is structured as: Negotiated F&A Rate. This rate must be applied correctly to the modified total direct cost base according to your institution's negotiated rate agreement or the flat rate set by the funder.

3. Step-by-Step Budget Justification Protocol

The budget justification (or budget narrative) is a critical component of the application reviewed by both financial auditors and peer reviewers. To draft a compliant narrative:

• Provide granular detail: Do not use lump sums. Break down personnel costs by calendar months or percentage of effort.
• Demonstrate direct linkage: For every cost, explain how it supports a specific task or objective in the research plan for Computer Science & AI.
• Cite institutional policies: Reference verified institutional rates for fringe benefits, travel mileage, and indirect cost bases to validate your numbers.
• Verify supplier quotes: For major equipment purchases or specialized laboratory assays, upload or reference formal vendor quotes.

Pre-Award Framework, Cost Sharing & Post-Award Governance

Pre-award research offices supporting grant development and pre-award grant management for NIH awards in Computer Science & AI must evaluate all eligible direct lines early in the application process. Unlike discretionary block grants given directly to departments, these funds are administered as categorical grants restricted to specified scientific deliverables under NIH rules. The study's grant proposal timeline must allow sufficient room for internal sign-off, subcontractor approvals, and the formal clearance of any required matching funds or cost sharing on grants. Once an award is finalized, robust post-award grant management takes over, requiring the immediate setup of a legally binding subaward agreement research with partner universities. Under active guidelines, project teams must submit formal effort certification research audits, enabling the PI to track personnel hours during collaborative team science research in Computer Science & AI.

4. Frequently Asked Questions