fellowship-application

name	fellowship-application
description	Apply for predoctoral fellowships (Hertz, NDSEG, NSF GRFP, Ford, AAAS Mass Media) and the Packard postdoctoral fellowship. Covers personal statement, research statement, transcript curation, grad school list, referee coordination, and submission timing. Use when applying for a major external fellowship as a PhD applicant or early postdoc.
license	MIT

Pre-Doctoral and Postdoctoral Fellowship Application

Hard rules

1. No fabricated citations. Every cited work must resolve to a verifiable
2. No claim without provenance. Every quantitative or factual claim
3. No silent failure. Every script invocation, API call, or tool use must declare its exit status and what to do on non-zero. A skill that silently swallows errors is a violation.

When to use

• Applying for a predoctoral fellowship as an undergraduate or early graduate student.
• Applying for a postdoctoral fellowship in the Packard, Pew, Searle, or similar private-foundation cohort.
• Building a master application that can be tailored to multiple fellowships with shared materials.
• Coordinating referees and timing across overlapping deadlines.
• Writing the personal statement and research statement required by major US fellowships.

When NOT to use

• Internal university fellowships — these are typically shorter and more institution-specific; consult your department's graduate office.
• Federal grants for senior researchers (NIH R01, NSF CAREER) — see ors-research-grants-nih-r01 and ors-research-grants-nsf-standard.
• Industry fellowships (some companies run 1-2 year programs) — these follow industry hiring norms; see ors-career-navigation-industry-transition.

Prerequisites

• A current CV (see ors-career-navigation-academic-cv).
• Transcripts (official and unofficial; most fellowships accept unofficial for application, official upon acceptance).
• A list of 3-5 potential referees: usually 1-2 research supervisors, 1-2 faculty who can speak to research potential, and optionally 1 who can speak to broader impacts.
• 1-2 page research summary that can be expanded into a full research statement.
• Knowledge of each fellowship's specific requirements (read the current solicitation — programs change year to year).
• A grad school list (for predoctoral applications): 5-10 programs, ranked by fit, deadline, and faculty overlap.

Core workflow

1. Build a fellowship matrix

Different fellowships target different populations and require different materials. Build a matrix:

Fellowship	Audience	Stipend / amount	Tenure	Citizenship	Deadline	Personal statement	Research statement	Transcripts	Refs
Hertz	1st-year PhD (applied math, applied physics, applied engineering, comp bio, computer science, earth sciences)	Stipend + cost-of-education allowance for up to 5 years	Up to 5 years	US citizen or permanent resident; open to those attending US schools	October (verify current year)	Yes	Yes	Yes	3
NDSEG	1st-3rd year PhD (DoD-relevant STEM)	Full stipend + tuition for up to 3 years	Up to 3 years	US citizen or national	October (verify)	Yes	Yes	Yes	3
NSF GRFP	1st-2nd year grad (research-based master's or PhD) or undergrads	$37k stipend + tuition allowance for 3 years	3 years	US citizen, national, or permanent resident	October (life sciences, engineering); late October / early November (verify current cycle)	Yes (2 pages)	Yes (2 pages)	Yes	3
Ford	Predoctoral, dissertation, postdoctoral; US-based	Predoc: stipend + tuition; Dissertation: stipend only; Postdoc: stipend only	Various	US citizen, national, permanent resident, or "non-rolling" eligibility for some (verify current year)	Various; check NSF portal	Yes	Yes	Yes	3
AAAS Mass Media	Graduate-student scientists placing in newsrooms	Stipend for summer	10 weeks	US-based graduate students	January (verify)	Yes (essay-style)	No (essay)	Yes	2
Packard (postdoc)	Early-career faculty-eligible; 22 US institutions nominate	$875k over 5 years (verify current amount)	5 years	US-based; institution nominates	Internal deadline Jan-Mar; national deadline May	Yes (research)	Yes	No	Multiple (institution-specific)

Verify each program's current year's details before relying on the table. Programs revise their amounts, deadlines, and eligibility year to year.

2. Build a master CV tailored to each fellowship

Most predoctoral fellowships are early-career, so the CV is short. The CV should highlight:

• Research experience (labs, projects, methods learned)
• Publications and preprints (a single first-author paper is a strong signal)
• Awards and honors (Dean's list, research awards, conference travel grants)
• Teaching and mentoring (TA positions, tutoring, outreach)
• Service and broader impacts (society membership, conference volunteering, science communication)
• Skills (programming, methods, instruments)

For each fellowship, the CV is usually submitted in a specific format (NSF GRFP uses the NSF-approved format; Hertz uses its own portal). Tailor the emphasis: for Hertz, lead with technical depth; for NSF GRFP, lead with broader impacts; for AAAS Mass Media, lead with writing and communication experience.

3. Personal statement

The personal statement is the most read document in a fellowship application. Length: typically 2 pages for NSF GRFP; 1-2 pages for Hertz; varies for Ford. Tone: personal, specific, forward-looking.

Structure (NSF GRFP is the most codified example):

1. Opening (1 short paragraph): who you are, your research question, your broader context.
2. Intellectual merit (1-2 paragraphs): your research experience, what you have done, what you have learned, what is novel. Be specific: methods, datasets, results.
3. Broader impacts (1-2 paragraphs): how your work benefits society, how you have contributed outside the lab (mentoring, outreach, teaching, science communication), how you will continue to contribute.
4. Future goals (1 paragraph): what you will do in graduate school, what you want to accomplish, what kind of career you are building toward.
5. Why this fellowship / why this graduate program (1 paragraph): the specific opportunity, the specific program, the specific mentor(s).

Style guidance:

• Personal and specific: avoid generic "I have always wanted to be a scientist" openings. Open with a specific moment, a specific question, a specific project.
• Concrete over abstract: "I built a Python pipeline that analyzed 1,200 metagenomes" is stronger than "I am interested in data science".
• Show, do not tell: instead of "I am a strong mentor", describe the mentee, the project, the outcome.
• Connect research and broader impacts: fellowships value integration, not parallel tracks.
• Have non-faculty readers review it: a parent, a friend, a non-scientist. If they can summarize your research and your goals, the statement is clear.

4. Research statement

The research statement is longer and more technical than the personal statement. Length: 2-3 pages for Hertz, 2 pages for NSF GRFP, varies for Ford.

Structure:

1. Research question (1 paragraph): what is the problem, what is the gap, what is your approach.
2. Background and prior work (1-2 paragraphs): what is known, what you have already done, what is novel about your contribution.
3. Proposed research (2-3 paragraphs): the specific aims, the methods, the expected outcomes, the alternative approaches. Specific aims are useful even in the research statement.
4. Timeline and feasibility (1 paragraph): what you will accomplish in the fellowship period; what resources you need; why the proposed institution and advisor are the right fit.
5. Why this fellowship (1 paragraph): the unique value of the fellowship, the community, the flexibility.

Style guidance:

• Figures welcome: 1-2 well-designed figures communicate a year of work in a single image.
• Be specific about methods: name techniques, software, datasets.
• Acknowledge limitations and alternatives: reviewers respect candidates who can see the failure modes of their own work.
• Connect to the proposed advisor's work: name 2-3 papers from the proposed group that are most relevant; explain how your project fits.

5. Graduate school list (predoctoral)

The graduate school list is strategic, not just a list of programs. Build it from three columns:

• Reach: top-5 programs, lower admit rate, ideal advisors
• Target: solid-fit programs, strong advisors in your area, high admit rate
• Safety: programs where you are clearly competitive

For each program, capture:

• School and department
• Potential advisors (3-5 names)
• Recent papers from the group
• Fit notes (why you, why them)
• Deadline and fee waiver policy
• Funding offer (stipend, tuition coverage, health insurance)

Apply to 5-10 programs. More than 10 dilutes effort; fewer than 5 leaves you exposed to a bad cycle.

6. Referee coordination

The referee plan is the most underestimated part of the application. Start 6-8 weeks before the deadline.

Selecting referees:

• 1-2 research supervisors (lab PI, summer supervisor)
• 1-2 faculty who can speak to research potential
• 1 who can speak to broader impacts (a mentor from outreach, teaching, or service)
• For postdocs: 3-4 senior researchers, including the postdoc advisor, a thesis committee member, a collaborator

Asking a referee:

Send a 1-paragraph email with:

• The fellowship name and deadline
• The URL of the program
• A copy of your CV
• A draft of your personal statement and research statement
• A list of the program prompts
• The submission link and any institution-specific requirements
• A deadline at least 1 week before the official deadline

Always ask if they can write a strong letter. A lukewarm letter is worse than no letter.

What to give the referee: the materials above plus a "letter cheat sheet": the program's evaluation criteria, the points you most want emphasized, the proposed research advisor and institution, and 2-3 specific examples of your work that they can cite.

7. Transcript curation

Most predoctoral applications ask for transcripts. The official transcript is the record; the personal statement and CV contextualize it. Some applications ask for a "transcript explanation" or "academic record" section — use this to:

• Briefly explain a poor semester or a gap
• Highlight research-heavy semesters
• Note non-coursework achievements (research, internships)
• Be honest; do not fabricate

A 4.0 GPA in a tough program is strong; a 3.5 in a tougher program with research and publications is also strong. The transcript is a data point, not a verdict.

8. Submission timing and revision cycles

Most predoctoral deadlines fall in October-December for fall start; some in January for spring. The Packard postdoctoral deadline is institution-internal in January-March and national in May.

Calendar (high-level):

Month	Action
6-9 months out	Identify target fellowships; build matrix; draft master materials
4-6 months out	Read each program solicitation; tailor materials; ask referees
2-3 months out	Draft full application per program; circulate to mentors for review
1 month out	Final revisions; check formatting; confirm referee letters submitted
2 weeks out	Submit; verify receipt; follow up with referees
After submission	Send thank-you emails to referees and mentors

Revision strategy:

• Write one strong draft and adapt it for each fellowship.
• Use a single source document (LaTeX or Markdown) with section files for each program.
• Have at least 2 mentors review each draft. Different mentors catch different things.
• Do not wait for the "perfect" version; submit a strong draft by the deadline.

9. Common applications for science writers (AAAS Mass Media)

The AAAS Mass Media fellowship places graduate-student scientists in newsrooms for the summer. The application is a portfolio:

• 3-5 writing samples (news articles, blog posts, op-eds)
• A personal statement (1-2 pages) on why you want to do science writing
• Transcripts
• 2 letters of reference (often from a science writer and a researcher)

Build a writing portfolio before you apply: pitch op-eds to local papers, write blog posts, contribute to a lab blog or institutional magazine, volunteer for science-communication organizations. The portfolio is the most evaluated component.

Code patterns

Fellowship application tracker

| Fellowship | Deadline | Status | Personal statement | Research statement | CV | Transcripts | Refs | Submitted |
|------------|----------|--------|--------------------|--------------------|----|--------------|------|-----------|
| Hertz | Oct X | Drafting | v2 | v1 | v3 | ordered | 2/3 confirmed | no |
| NSF GRFP | Oct X | Drafting | v2 | v1 | v3 | ordered | 2/3 confirmed | no |
| NDSEG | Oct X | Drafting | v2 | v1 | v3 | ordered | 2/3 confirmed | no |

Master application source layout

/Applications/Hertz/
  personal-statement.tex
  research-statement.tex
  cv.tex
  transcripts/
  referee-cheat-sheet.md

/Applications/NSF-GRFP/
  personal-statement.tex   (NSF-specific format)
  research-statement.tex   (NSF-specific format)
  cv.pdf
  transcripts/
  broader-impacts.md
  referee-cheat-sheet.md

Personal statement opening examples (avoid generic openings)

# Avoid
"I have always been fascinated by science. From a young age, I knew I
wanted to be a researcher..."

# Better
"In my second month in Dr. Garcia's lab, my qPCR plate failed twice
in a row. The third time, I changed the primer design and the data
finally made sense. That week I realized I was not just running
protocols — I was designing experiments. Three years later, I lead
the wet-lab component of a project..."

# Better still (with a specific question)
"Can a single-molecule measurement of a ribosome tell us why
translation slows in aging neurons? This is the question I have
spent the last two years learning to ask, and the question I want
to spend my PhD answering."

Research statement figure conventions

• One figure per page max; usually 1-2 figures total.
• Use the same fonts and color palette as your main paper.
• Label axes clearly; do not assume the reader knows your field's conventions.
• Caption: 2-4 sentences that stand alone, summarizing what the figure shows.

Referee request email

Subject: Letter of recommendation request for [Fellowship]

Dear [Referee],

I am applying to [Fellowship] this fall. The deadline is [date].
The program description is here: [URL].

I have attached my CV, draft personal statement, and draft research
statement. The program asks for [3] letters; the most relevant
points the committee evaluates are:
- [Criterion 1]
- [Criterion 2]
- [Criterion 3]

If you can write a strong letter on my behalf, I would be grateful.
The submission portal is [URL]; I will send the link to your portal
account by [date 1 week before deadline].

If you are unable to write a strong letter, please let me know
honestly — I will find an alternative referee.

Thank you for considering this request.

[Your name]

Common pitfalls

Pitfall	Why it fails	Fix
Generic personal statement	Reviewer cannot tell which applicant wrote it	Open with a specific moment, project, or question
Personal statement that lists publications	The CV lists publications; the statement is for narrative	Tell the story behind the work
Research statement as a paper draft	Reads as truncated, not as a plan	Restructure around specific aims, methods, timeline
Failing to address broader impacts	NSF GRFP and others weigh this heavily; omission signals poor fit	Allocate 30-40% of personal statement to impacts
Applying to 1-2 programs	Concentrates risk; no leverage if rejected	Apply to 5-10; cross-cut by fit, deadline, and competitiveness
Asking referees too late	Refusals; rushed letters	Ask 6-8 weeks before deadline
Asking a stranger for a letter	Generic letter; low impact	Use referees who know your work in detail
Submitting a fellowship application with a "B" CV	The CV is a single document; it should be "A"	Polish CV once; use it across all applications
Inconsistent research summary across materials	Personal statement, research statement, and CV disagree	Maintain a single source of truth; tail only the emphasis
Missing institution-internal deadlines	Packard-style programs require institutional nomination	Identify the internal deadline; submit to the institutional coordinator first
Ignoring the citizenship / eligibility rules	Wasted application; disqualification	Check eligibility before drafting
Using AI-generated text without review	Reviewers can tell; institutional policies vary	Use AI for editing and brainstorming; write the substance yourself
Forgetting to send thank-you notes to referees	Missed relationship-building	Send a short note the day after submission
Skipping the writing portfolio for AAAS Mass Media	The portfolio is the application	Build the portfolio months before the deadline
Treating Ford as a "stretch" application only	Ford has multiple levels; check your eligibility year	Apply at the right level (predoctoral, dissertation, postdoctoral)

Validation

A complete fellowship-application package satisfies:

• Fellowship matrix built; 3-6 target fellowships identified
• Master CV polished; tailored to each application
• Personal statement (master draft) with intellectual merit + broader impacts + future goals
• Research statement (master draft) with question, background, proposed work, timeline, fit
• For science writing: 3-5 writing samples in the portfolio
• Transcripts ordered or available (official and unofficial)
• 3-4 referees identified and confirmed; materials sent 6-8 weeks before deadline
• Referee cheat sheet prepared for each referee
• Application submitted at least 48 hours before the deadline
• Confirmation emails saved; portal access verified
• Thank-you notes ready for referees and mentors
• Calendar of related deadlines (e.g., grad school, other fellowships) maintained

Open alternatives

Commercial / proprietary	Open equivalent	Trade-offs
Application portals (NSF FastLane, Research.gov, Hertz portal)	Each program's official portal is the only path	Portals are mandated; you must use them
Paid editing services	A trusted mentor in your field; university writing center	Mentor review is more substantive and tailored
Chegg / Course Hero for transcript explanation	Direct conversation with the registrar and your advisor	Both have limitations; honesty is the best policy
LinkedIn Premium for referee research	Institutional websites, Google Scholar, ORCID, OpenAlex	Public sources are usually sufficient

References

• Hertz Foundation fellowship: https://www.hertzfoundation.org/
• NSF Graduate Research Fellowship Program (GRFP): https://www.nsfgrfp.org/
• NDSEG (National Defense Science and Engineering Graduate Fellowship): https://ndseg.sysplus.com/
• Ford Foundation Fellowship Programs: https://www.nsf.gov/funding/programs/ford-fellowships
• AAAS Mass Media Science and Engineering Fellowship: https://www.aaas.org/programs/mass-media-fellowship
• Packard Foundation Fellowship for Science and Engineering: https://www.packard.org/funding-for-researchers/fellowships-for-science-and-engineering/
• NSF PAPPG 24-1 (general proposal structure reference)
• National Academies "Advisor, Mentor, and Advocate" resources (public)
• Public graduate-school admissions guides at major US universities

Related Skills

• ors-career-navigation-academic-cv — CV source
• ors-career-navigation-tenure-dossier — long-term career planning referenced in the personal statement
• ors-career-navigation-faculty-interview — if a fellowship unlocks a specific graduate program or position
• ors-research-grants-nsf-standard — for senior fellowships and grant applications
• ors-scientific-writing — research statement and personal statement structure
• ors-mentorship-teaching-ors-mentorship-goal-setting — for the mentoring / future-goals content
• ors-tailored-resume-generator — for the CV tailoring

Changelog

• 1.0.0 (2026-06-10): Initial adaptation by Pradyumna Jayaram. Compiled from public Hertz, NSF GRFP, NDSEG, Ford, AAAS, and Packard program descriptions; cross-referenced to CV and grants skills.

Cross-references

Other skills in this category:

• academic-cv
• faculty-interview
• industry-transition
• negotiation
• tenure-dossier