GRFP_logo.png

NSF GRFP - tips/advice

Jessica Yin - Awarded in 2020

I was extremely fortunate to be awarded the NSF Graduate Research Fellowship. I applied in my senior fall for the subject area of Robotics & Computer Vision. Of course, I credit a lot of my application’s success to a lot of help from professors, graduate students, and online tutorials/pages from previous successful applicants. I hope to add to the online knowledge base, so hopefully you’ll find this helpful if you’re applying!

Links to my proposal, personal statement, and ratings from NSF reviewers.

Online Resources

I don’t want to be too repetitive with a lot of online resources out there already, so I won’t go over most of the things that these resources already include. Instead, I’ll link them here below and I’ll focus on writing about what I think is particularly important or what I haven’t seen as frequently.

In-Person Resources

  1. Ask your professors to help you! ESPECIALLY your letter writers! I cannot emphasize this enough. You are writing a grant proposal, and your professors are literally professional grant proposal writers. They have been doing this successfully for YEARS, both as applicants and reviewers, maybe even for the NSF GRFP. Even if they weren’t NSF GRFP reviewers, the NSF GRFP reviewers are almost always professors anyways. I went through about 6 drafts with the two professors who also wrote my letters for grad school and the fellowship applications (my other letter writer was one of my internship supervisors). And although you probably have a good relationship with the professors you’re asking to write letters of recommendation for you so it’s easier to ask them for help, it is also extremely valuable for the professors to directly discuss your NSF application in their letters of recommendation. I personally know that one of the professors who helped me revise my application also added his own reasoning as to why my proposal should be funded, which adds a ton of credibility to the application. Remember: they can write however much they want in the letter, while we have strict page limits.

  2. Ask graduate students and postdocs to help you, particularly labmates if you are planning to continue your current research area in graduate school. They typically have a more flexible schedule and a quicker turnaround time than professors do (although they are still doing you a huge favor and should be accordingly thanked). Pretty much everyone has been in academic research longer than you have as a full-time student or employee, and will have valuable insight and suggestions. I recommend having the graduate students and postdocs do the “first-pass” editing and revisions with you, and later sending the more polished versions to the professors. However, this assumes that you started early enough for multiple iterations with both graduate students/postdocs AND professors (who need more time to help you).

  3. Attend virtual/in-person writing workshops or schedule an appointment with a writing tutor at your school. I would recommend this only after you have a solid idea of the technical content, but I found this extremely helpful for overcoming writer’s block. Especially with one-on-one peer advising, being forced to discuss and describe my ideas out loud to a person not familiar with my field helped me organize my thoughts and select the essential details. I would highly recommend this for the personal statement, at the very least, since that’s the less technical part of the application and easier to get good general writing advice for.

In general, I hope you get the idea that you should ask for help, early and often. Don’t be afraid of multiple iterations and drafts, or asking people for help. Everyone I’ve asked for help has been extremely nice and happy to look it over!

Things I Wish I knew earlier

  1. You don’t have to submit to the subject that your degree is in. I was a mechanical engineering major and I didn’t do a double major or even a minor in Robotics (I know, despite attending CMU). Although I did submit to the Robotics & Computer Vision category, which looking back was the only topic that would’ve made sense anyway, I was still very nervous about it. I suffered (and still do) from a nice case of imposter syndrome, worrying that despite my 3 years of research experience in soft robotics, the expert reviewers would see through my lack of a formal robotics education. However, with my proposal that focused on closed-loop control, sensors, rudimentary computer vision, autonomy, etc. it would have been out of place in Mechanical Engineering. Choose the subject area that is the best fit for your proposal, which is not necessarily the same as your undergraduate studies.

  2. You can submit a proposal on research that is currently in progress. Although this might have been somewhat of a unique case, my senior fall was extremely hectic since I also started and finished a research project for a paper in 2.5 months that semester. I had a very difficult time coming up with another idea since I was already so entrenched in the rapid pace of my research at the time. However, because I knew my project and the technical content pretty well already, it was easier to write the proposal. Remember, what NSF GRFP really looks for is: Can you write a good grant proposal? Do you have the potential to conduct high-impact scientific research in your future career? At first, I was a little uncomfortable because I thought it had to be a completely new idea specifically for the NSF GRFP, but that misses the point of the program. And in the end, I was at such an early stage in my research at the time when I wrote my NSF application that my paper ended up different than the proposal anyway.

  3. You need to start early. This is not because of the pace of your own writing, but for the timeline of everyone you will ask for help before you submit. I started about a month before the deadline, which just made it more stressful for me as I tried to give ample time to all the professors and graduate students I asked for help.

  4. Be involved with STEM outreach activities. I do hesitate a bit with this point since I really don’t like recommending doing certain activities because of how it will look in your CV/resume. However, it is an important part of your application, since NSF also looks at how you personally have helped get more people interested in STEM. You can’t just be a technical genius! Ideally, you are already very passionate about this and have been continually involved with STEM outreach throughout your college career. I remember being thankful that I had already participated in several STEM outreach activities before I had even thought about graduate school or applying for NSF, so I hope that is also a lucky coincidence for you.