GRFP, NDSEG, Hertz: Apply to All Three in One Season

GRFP, NDSEG, Hertz: Apply to All Three in One Season

Most U.S.-eligible STEM PhD applicants treat the NSF Graduate Research Fellowship, the National Defense Science and Engineering Graduate (NDSEG) Fellowship, and the Hertz Fellowship as a choice. The Hertz Foundation does not. Asked in its own application FAQ whether a candidate holding another institution's offer must turn it down, the Foundation answers:

"No, you don't need to turn down the other offer. We coordinate with a variety of institutions, including the National Science Foundation, National Institutes of Health, and the National Defense Science and Engineering Graduate Fellowship, to make sure you can take advantage of both offers and receive the maximum possible compensation."

That is a sponsor stating in print that the multi-award portfolio is a coordinated, expected outcome. The three applications also overlap across the same September-to-October stretch — NDSEG's window is the longest of them, August 3 to October 30 — and they fund the same person for the same graduate years. So for anyone eligible for more than one, the question is not which — it is how, because the three applications look interchangeable and are not. Below: the deadline ladder, the eligibility deltas that quietly disqualify people, and which raw material you can reuse before reuse is what stops an application from moving forward.

Quick answer

• NDSEG FY2027: August 3 through October 30, 2026, 5 p.m. Eastern.
• Hertz 2027: the application "will open on August 31, 2026," and the Foundation's own timeline lists a late-October 2026 deadline. It has not published a day of the month; do not assume one.
• NSF GRFP FY2027: not posted yet. NSF 25-547 is still the listed solicitation. GRFP deadlines are always staggered across four dates by field, and they moved between mid-October and mid-November across the last two cycles. Watch the NSF page; do not roll a date forward.
• The three do not share an essay contract. GRFP requires named Intellectual Merit and Broader Impacts headings; NDSEG requires redaction and treats a leak as grounds to stop the application from moving forward; Hertz requires no research proposal at all.
• Reuse the evidence. Rewrite the document. Every time.

One season, three deadlines

Everything below was checked on the sponsors' own pages on August 11, 2026.

The empty GRFP cells are the honest answer, and the reason to start early. NSF posted the FY2026 solicitation on September 26, 2025, with field-staggered deadlines November 10–14, 2025. The cycle before, NSF 24-591 posted July 12, 2024, with deadlines October 15–18, 2024. Two months of movement between cycles: a pattern, not a prediction. You cannot sequence around a date you do not have, but you can build the shared material in August and September and be ready either way.

The reference bottleneck is where the three diverge invisibly

The essays get the attention; the recommenders are the actual constraint, and the three sponsors ask three different things of the same busy professor:

• GRFP: list 3–5 writers; the top 3 by your own ranking are reviewed; if fewer than two letters arrive the application is not reviewed at all. Letters have their own deadline.
• NDSEG: up to 3 references, "highly recommended but not required," submitted on a form via a system link — and "Letters on letterhead will NOT be accepted." A professor who writes you a beautiful two-page letter has produced something NDSEG cannot take.
• Hertz: four recommenders completing an evaluation form rather than a letter, asked for "notable examples of personal creativity," your strengths and weaknesses at your current training stage, your potential as an applied scientist or engineer, and a peer comparison.

Warn your writers about all three formats in August, in one message. What NSF GRFP reference letters are scored on covers the GRFP side, including the all-merit, no-Broader-Impacts failure pattern; if a writer asks you to draft your own, see where that line sits.

Eligibility deltas at a glance

The most useful line in that table is the citizenship row. Permanent residents are eligible for GRFP and Hertz and ineligible for NDSEG. If that is your status, your fall has two applications in it, not three — better learned in August than in October.

One repeat rule gets misread constantly: GRFP allows one application per annual competition, and graduate students and joint-degree students get one lifetime application. Seniors and bachelor's holders without graduate enrollment may apply more than once before entering graduate school; the grad-student attempt is a one-shot.

On holding two awards, the Hertz FAQ describes coordination with NSF, NIH, and NDSEG, while NDSEG's eligibility page says fellows may not hold competing fellowship funding simultaneously. Those get resolved sponsor-to-sponsor at offer time. It is a question you only get to have because you applied to more than one — not a reason to skip an application.

One published data point for scale: Hertz announced its 2026 class after what the Foundation called the highest number of applications it had ever received — 1,493 applications, 119 interviewed, 54 second-round finalists, 19 fellows — for an award valued at up to $250,000 over as many as five years. Nineteen out of 1,493 is about 1.27%, and that percentage is our own calculation from the Foundation's two published counts for one past cycle: not a Foundation-published selection rate, and not a forecast for 2027.

What transfers, and what must not

What genuinely transfers is raw material, never prose:

• The research you actually did — the question, your specific contribution, the method, the decision points, what the result changed.
• Your trajectory: how the interest formed, the turns you took, what you chose.
• Outreach, teaching, mentoring, and service — the activities themselves.
• Your reasons for the next five years: the direction, the problem, the training you still need.
• The briefing packet for your recommenders: three different forms, one set of facts about you.

What does not transfer is every sentence you wrote. Here is the contract each sponsor actually enforces:

1. GRFP: the headings are the contract

NSF 25-547 requires that both statements carry "separate sections, under individual separate headings, on 'Intellectual Merit' and 'Broader Impacts'," and says applications lacking the distinct headings "will be returned without review." That requirement is peculiar to NSF, and it makes GRFP the only one of the three where your outreach, teaching, and mentoring work has a required home on the page.

The corollary runs the other way too: NSF 25-547 contains no anonymization requirement for the statements (it does require redacting personal identifiers from transcripts). Naming your university, advisor, lab, and collaborators is ordinary GRFP practice — which is exactly what makes the GRFP draft dangerous later. The FY2027 GRFP cycle guide covers the solicitation mechanics.

2. NDSEG: redaction is enforced

The NDSEG FAQ defines the rule directly:

"Redaction is defined as no inclusion of any identifying information, such as your or any person(s) names, organization/institution names, or specific details that could tie the application back to you and any institution/organization or person. Failure to follow this guidance will prevent your application from moving forward in the evaluation process."

Read that against the previous section and the consequence is blunt: the GRFP draft you are proudest of is the most dangerous file on your desktop during NDSEG week. Everything that makes it good — the named lab, the named PI, the named university, the named summer research program — is exactly what NDSEG requires you to remove.

NDSEG is also not one essay. It is two 500-word boxes with two different jobs: one on your short- and long-term professional goals, how they developed, the foundation you have laid, and how the fellowship fits; the other on your personal story and the journey to where you are today. A polished GRFP personal statement cannot be cut in half to fill both.

The rescue is the useful part: redaction is not vagueness. "A large public research university" and "my advisor" are fine. What must survive is the anonymized substance — what you did, what you chose, which method you used, what you found, what you learned. Strip the names and the specifics together and you are left with a résumé of redacted nouns and no person in it.

The same discipline governs the proposal, with its dual original-and-redacted upload; both documents are covered in the NDSEG essays guide for FY2027 and the NDSEG research proposal and redaction guide. One nuance worth carrying: the FAQ says applicants are "NOT required to fulfill any of the requirements put forth by the DoD Strategy Documents." You align your research to DoD priorities. You do not satisfy a solicitation, and you do not contact anyone at DoD.

3. Hertz: there is no research proposal, and writing one is an active mistake

The Hertz FAQ is unambiguous:

"The fellowship is intended to fund individual students, not a specific project. There is no requirement to submit a proposed research project as part of the application process. However, the application does provide the opportunity to summarize current and prior research experiences as well as interests you intend to pursue."

A GRFP two-page research plan compressed into a Hertz short answer is a strong answer to a question nobody asked. What the Foundation wants instead is the person. Of the personal essay the Foundation says: "We want you to reveal your creativity beyond your research. This could be your favorite extracurricular activity, a leadership experience, or anything else that gives us a better sense of what makes you who you are." The recommender form then asks four separate people for "notable examples of personal creativity." When a sponsor asks four different people for evidence of the same trait, that trait is the rubric.

The length culture differs too: Hertz's published ranges are suggestions, because "reviewers are less concerned with the word length and more so about the applicant's ability to adequately address the prompt" — the opposite of NSF's enforced page caps and NDSEG's per-box ceiling. Read the 2027 prompts in the portal when it opens, not in a secondhand list.

The three rules

Rewrite, never paste. Reuse the experience; retype the sentences. A paragraph written to satisfy a Broader Impacts heading will not perform an NDSEG personal-journey box, and it may carry a name that stops your NDSEG application from moving forward.

Redact by document, not by habit. Keep one clean, fully-named master file and produce the redacted versions from it. Never edit toward NDSEG inside the file you will submit to NSF, and never assume a "cleaned" file is clean — reread it for the school name in the fourth paragraph.

Answer the question that was asked. All three sponsors published what they want. The failure mode across every one of them is submitting a strong answer to a different sponsor's question.

The same discipline applies inside one application: NDSEG's two boxes each have to do their own work, and Hertz's short essays should not restate the personal essay. For the general version, see scholarship essay examples by prompt type and how to write a scholarship essay that fits the award.

The fourth award: DOE CSGF

The Department of Energy Computational Science Graduate Fellowship is the fourth award many of these applicants will want, and as of August 11, 2026 its next cycle has not posted — the application page offers only a notify-me signup, and the public FAQ still describes the prior cycle. The program historically runs on a different rhythm, a fall opening and a January deadline, which is why CSGF sits outside the September-to-October crunch. Check the Krell Institute's official page rather than rolling last year's dates forward.

How one set of material becomes three applications

• August: build the master file — fully named, everything in it: projects, contributions, methods, decisions, results, outreach, teaching, and the reasons for the next five years. It is never submitted anywhere.
• Late August and early September: NDSEG is already open. Produce the two redacted 500-word boxes and the redacted proposal from the master, keeping the original proposal for the dual upload.
• August 31: Hertz opens. Write to creativity, independence, and applied intent — not to a project.
• September and October: GRFP, once it posts, with the Intellectual Merit and Broader Impacts headings in the first outline rather than retrofitted later.
• Throughout: one briefing packet, three different forms, everyone warned early.

Where a review fits

NDSEG's own guidance tells applicants to "have an Advisor review your Proposal, Personal Statement/Application." That stands, and nothing here replaces it. A rubric review takes a pass before your advisor's limited time: your own draft read against that award's criteria, one document at a time, so the version your advisor sees is not the first one.

Four public rubrics cover the GRFP and NDSEG documents today — NDSEG application essays, NDSEG research proposal, GRFP personal statement, and GRFP research plan — plus the GRFP reference-letter rubric for writers, all on the scholarship rubric shelf. Entry points are the NSF GRFP application review and scholarship essay review pages.

Two boundaries, stated plainly. A text review reads the text you paste; it can flag a school name, a person's name, or a program name sitting in your NDSEG draft, but it cannot certify that a document is anonymous. Whether an ordinary detail — an unusual instrument, a specific field site, a one-of-a-kind collaboration — identifies you to a reviewer in your subfield is a judgment only you and your advisor can make. And no review determines eligibility, verifies facts outside the text, or predicts an award; how the scholarship review works sets out the full boundary. Two free Quick Reviews are available each day; your first Full Review is $5, once per account; Pro is $50 for ten Full Reviews.

The thesis, once more

Same season, same person, three different contracts. The Hertz Foundation says in print that it coordinates with NSF and NDSEG so a winner of two can hold both, which makes the portfolio the rational play for anyone eligible for more than one. Reuse what you did; rewrite what you say. And put the ladder on the calendar now: NDSEG closes October 30, 2026 at 5 p.m. Eastern, Hertz opens August 31, 2026 and closes in late October, and GRFP's FY2027 dates land whenever NSF publishes them.

---

Sources

All checked August 11, 2026.

• NDSEG — program portal, How to Apply, Application, Evaluation, and Award, Application FAQ, and Eligibility.
• Hertz Foundation — Apply, Application FAQ, Who Can Apply, and the 2026 fellows announcement.
• NSF — Graduate Research Fellowship Program, solicitation NSF 25-547, and archived NSF 24-591.
• Krell Institute — DOE CSGF application access.