Foundation grant writing with every section inside the funder's limit, every figure cited, and a named person signing each part
Proposals drafted to each funder's limits and rubric, every figure cited to its source.
See one case, screen by screen ↓Foundation grant writing is turning a funder's call into a proposal: each section inside the funder's limit, every figure drawn from the institution's own data, a budget it will honor, and a draft that answers the funder's rubric. Foundation Grant Writing puts seven agents beside the grant writer for that work, and a named person accepts every text and signs every figure, budget and submission.
AI in education: a buyer's guide to agentic AI for colleges, universities and school districts →
Updated
Foundation grant writing at a college foundation means a new call every few days, and every funder has its own: its own sections and character limits, its own priorities, its own scoring rubric and its own reporting terms. Many corporate and family foundations accept only a few proposals from one institution a year, so each one has to count.
Today a grant writer copies program facts from old proposals, emails the dean and waits on institutional research for outcomes, rebuilds the budget from last year's salary file and counts characters in the funder's portal on the deadline day. Sections run over the limit, a count small enough to identify a student slips in, the budget narrative disagrees with the budget, and the sustainability plan says the college will seek other funding, which funders score low. A year later, the funder report is written from memory.
Estimated split of a typical proposal's business days, from the funder's call to an approved proposal, by hand and with the agents. The 'before' column totals the sample office's median of 11; the 'after' column totals the 5-day target. Our arithmetic, kept modest: the people who confirm, release and sign keep most of their time.
Seven agents read the call, gather the figures, build the budget and score the draft; named people sign each part.
Proposal FR-27-018 asks the Corwell Energy Foundation's Workforce Pathways Fund for $750,000 over three years to expand Brandvale Community College's electrical lineworker program. It is due Friday, November 13 at 5:00 PM. Here is what happens to it, screen by screen, in the working solution.
The Guideline agent read the six-page call and set one block per section, each sized to its character limit, with the funder's rubric down the right. Statement of need sits at 2,410 / 2,500 with 90 left; Outcomes and measures is hatched, 1,352 / 1,250 and 102 over.
9 business days to the deadline.
Jonah clicks 87% in the Statement of need. It opens its source: 54 of 62 completers of the 2024 cohort employed in the field within six months, from the state wage record match run by the Office of Institutional Research, as of September 30, 2026, released by Dr. Priya Natarajan on October 30. The $31.40 median hourly wage beside it carries its own source.
The 2024 counts of women and veteran completers are fewer than 10 each, so the Fact agent withholds them and offers three-cohort roll-ups instead: 13 of 178 and 19 of 178 completers across 2022–2024, which Institutional Research released. Race and ethnicity for 2024 had 4 of 6 groups under 10 and was not used.
Aggregates only. Minimum cell size 10. Every figure carries its source.
The Reviewer agent scores version 6 against the call's own rubric and weights: 76 of 100, below the site-visit line of 80. Need and alignment earns 5 of 5, program design 4, outcomes 4, evaluation 3, budget 4, and sustainability 2 of 5. Each score shows its reason, what would raise it and the call's own words.
Advisory, never blocking.
The draft only said the college would look for further support. The Drafting agent proposes, as tracked sentences, that the second lead instructor moves to the general fund in fiscal year 2031 at $112,600, by Board of Trustees resolution 2026-31 of June 18, 2026. The section goes from 403 to 1,087 of 1,250 characters.
§5 of the call: "A general intention to seek other funding is not sufficient."
Jonah accepts the plan. Sustainability now scores 4 of 5 and the predicted total rises from 76 to 82 against the line of 80, in version 7. Outcomes is still 102 over its limit and the budget narrative still disagrees with the budget, so both stay on his list.
The Budget agent read fiscal year 2027 salary and fringe rates from the ERP on October 28: a lead lineworker instructor at $78,400 with fringe at 34.2%, 720 part-time hours at $52.00, twelve training poles for $38,329. Indirect costs are $75,000, 10% of the request, the Fund's limit. The narrative said 1,080 part-time hours, so the agent proposes correcting the narrative, not the budget.
The chain names each signer: the grant writer, the Dean, Institutional Research, the Controller and the Foundation Executive Director. Four blocking checks are open: Outcomes over its limit, the narrative's 1,080 hours against 720 in the budget, a changed figure of 21 to 22 dual-enrollment students, and the unsigned budget. Approval stays closed until all four are clear.
Jonah accepts both smaller fixes, Outcomes tightened to 1,200 characters, and Dr. Alma Vasquez-Reid, Dean, confirms the changed figure. Gloria Benedetti then signs the budget: $750,000 in three years, $75,000 indirect at 10%, the college's share of $124,236 and the board-approved $112,600 line. Name, time, reason and evidence go to the audit trail.
Celeste approves submission of the frozen version with code APR-02, with one advisory still open on Evaluation, and Jonah submits through the funder's portal. The Submission agent re-checks every limit in the portal's format and files receipt CEF-WPF-2027-0418.
The same proposal, seen by the people who write, release, sign and audit it: what their day looked like, and what it looks like now.
Seven agents read the call, gather the figures, build the budget and score the draft; named people sign each part.
Reads the funder's call and sets the sections, character limits, priorities, rubric, attachments and dates, citing the paragraph for each.
Pulls enrollment, completion and wage outcomes from institutional research, the student information system and the state wage record match, with source, cohort, n and date for every figure.
Writes each section in the college's voice inside its limit, from released figures and approved sources, and proposes every change as a tracked edit with its reason.
Builds the budget from ERP salary schedules and fringe tables, applies the funder's indirect limit and checks every number in the budget narrative against the lines.
Scores each criterion 1 to 5 with the call's own weights, explains each score with the passage that earned it, and names what would raise it.
Tracks the sign-off chain, holds approval while a blocking check is open, then freezes the approved version, re-checks it in the portal's format and files the receipt.
Drafts each interim and final funder report from the same fact ledger and the grants system, promised against delivered.
Accept, confirm, sign, approve. The agents propose; a named person decides.
The assistant answers from the proposal, the call and the fact ledger, with a citation on every answer, and can take you to the section it is talking about.
Why did Sustainability score 2 of 5?
The call asks how positions paid by the grant will be funded afterwards, and says a general intention to seek other funding is not sufficient. The draft only says the college will look for further support. The Board approved moving the position to the general fund in fiscal year 2031, resolution 2026-31, so the Drafting agent can write it from that line. That would lift the total from 76 to 82, above the site-visit line of 80.
Where does the 87% placement come from?
From the state wage record match run by the Office of Institutional Research: 54 of 62 completers of the 2024 cohort were employed in the field within six months. The median hourly wage of those 54 is $31.40. Dr. Priya Natarajan released both on October 30.
Can we say how many women completed in 2024?
Not for 2024 alone: the count is under 10, and the college never reports a cell that small. The Fact agent rolled it up across the 2022–2024 cohorts: 13 of 178 completers, 7%, which Institutional Research released. The call also asks for aggregate data only.
Does the budget match the narrative?
All but one number. The narrative says 1,080 part-time hours; the budget line is 720 hours at $52.00, $37,440. 1,080 hours would cost $56,160 and push year 1 over the request, so the Budget agent proposes correcting the narrative, not the budget.
13 screens from the working solution, on its sample data. Pick one to see it large.
The limits, the FERPA rules, the approvers and the retention every proposal runs on are settings; the ones policy requires are locked, and say why.
| Setting | Default | Choose from |
|---|---|---|
| How limits are counted | Characters with spaces | Characters without spaces · Words · Pages, 12-point, single-spaced; set per funder from the call |
| Warn when a section passes | 95% of the limit | 90% · 95% · 100% of the limit |
| Minimum cell size | 10 | Stepped; any count below it is withheld from every proposal and report |
| When a cell is too small | Combine up to three cohorts | Show "fewer than 10" only · Leave the figure out |
| Budget signed by | Controller | Controller · Chief Business Officer |
| Submission approved by | Foundation Executive Director | Asks over $1,000,000 also need the President |
| Submitted versions and their evidence | Kept 7 years after the grant closes | 5 · 7 · 10 years; drafts never submitted removed after 3 years |
| Escalate a late signature | After 1 business day | Same day · After 1 business day · After 2 business days, to the Foundation Executive Director |
Every figure is labelled: a target the solution is built to, an estimate, a typical published result, or a proven one.
Sources: Giving USA 2025 (US foundations gave $109.81 billion in 2024; education received $88.32 billion, up 13.2%) · FERPA, 34 CFR Part 99, and de-identified records, 34 CFR 99.31(b)(1) · "target" = design goal, measured in the live solution · "demo" = seen in the working solution, on its sample data · "typical" = published figures · "estimated" = our arithmetic, kept modest. Brandvale Community College, its foundation, the funders and the people named on this page are fictional: characters and sample data in the working solution.
Foundation grant writing is turning a funder's call into a proposal: each section inside the funder's limit, every figure drawn from the institution's own data, a budget it will honor, and a draft that answers the funder's rubric. Foundation Grant Writing puts seven agents beside the grant writer for that work, and a named person accepts every text and signs every figure, budget and submission.
The Drafting agent writes each section inside its limit, but only from figures in the fact ledger and sources approved for external use, such as a board resolution, and every change is a tracked edit with its reason. It never accepts its own text: the grant writer accepts, edits first or rejects with a reason code, and can undo.
The budget narrative explains the budget in words. The Budget agent builds the budget from the ERP's salary schedules and fringe tables, then compares every number in the narrative with the lines. In the working solution it found a narrative saying 1,080 part-time hours against 720 hours at $52.00 in the budget, and proposed correcting the narrative. The Controller signs the budget.
Yes. The Reviewer agent scores each criterion 1 to 5 with the call's own weights, explains each score with the passage that earned it, and names what would raise it. The score is a prediction to decide where to spend effort, never a promise to the funder, and it never blocks approval.
The agents read aggregate views only; sending them student-level records is locked off. Any count under the minimum cell size, 10 by default, is withheld and offered as a roll-up across up to three cohorts, and Institutional Research releases every student figure before it can be used. Evidence packs export with staff contacts redacted.
In the working solution's sample office, the median from a funder's call to an approved draft was 11 business days in the last fiscal year. The design goal for Foundation Grant Writing is 5 business days from the call to an approved proposal, measured in the live solution.
Yes. The writer accepts or rejects every agent text, the dean confirms program facts, Institutional Research releases student figures, the Controller signs the budget and the Foundation Executive Director approves submission. Approval stays closed while any blocking check is open, and no agent submits on its own.
The Agentic Solution Engine builds and deploys it from your requirements and documents: your funder calls, past proposals, outcome definitions and rates. It goes live once every quality gate has passed, and we will walk you through it on one of your own calls first.
Every charge checked against the donor's own words, and idle payout put to work.
1-min demoProspect ResearchA cited brief before every visit in minutes, not days, each wealth figure confirmed twice.
1-min demoGift Acceptance & Donor Due DiligenceEvery large gift and naming decided on a cited file, with returnable gifts flagged early.
1-min demoWe'll run Foundation Grant Writing on one of your funder calls, with your own program data and rates.