Published October 28, 2026.
Email the person who would supervise you, about one specific piece of their work, with one ask and a one-page résumé. The student sends it. Follow up once, about a week later, on the same thread. If there is no reply, write to someone else, or apply to a program that already has mentors. A paragraph that could go to any lab is the one that gets skipped. The useful note is short enough that they can answer it in a few lines.
What to send
| Piece | What belongs there |
|---|---|
| Who | Your name, grade, school, and the city. One sentence. |
| Why them | One paper, poster, or project of theirs, and the part you actually read. |
| The ask | One task, or one short visit. Name the hours you can really do. |
| Proof | A one-page résumé. A grade in the relevant class beats a list of clubs. |
| Follow-up | One reply, a week later. Then a different person. |
Write to the graduate student or lab manager when the professor’s page says they run the day-to-day work. Professors forward a clear note. They rarely answer a request to be mentored in the abstract. Do not attach a five-page proposal for a lab you have not been invited into, and do not ask them to invent a project for you. If the school fair needs a measured question, the ISEF rules are the constraint. A mentor still has to agree before you touch their data, and the fair form has to say which part you did.
I would send three specific emails, not thirty copies of the same one. I would also apply to a program whose dates are already published, so a silent inbox is not the whole summer. Clark Scholars takes 12 students, pays a $750 stipend after the research report, and the 2027 application is due February 15. Fermilab’s PRISM internship pays $500 a week and requires an Illinois high school. Garcia’s summer program lists a June 24, 2026 start on its homepage. Which of those fits is in the mentors guide.
Reading the papers before writing the email
Professors can tell within a sentence whether a student read their work. Before any email goes out, we have the student read the abstract and introduction of one or two recent papers from the lab and write, in their own words, one question those papers raised. That question becomes the heart of the email.
Everything else stays short: who the student is, the skills they bring, a specific offer such as data cleaning or a literature review, and an attached résumé. Write to a few labs at a time, follow up once after about a week, and keep notes on every reply. If cold emails do not land, our guide to research summer programs lists structured options.
Questions families ask
Who should a high school student email?
The person who would actually supervise the work. That is often a graduate student or the lab manager, not only the professor whose name is on the lab. Read one paper or one project page of theirs first, and write about that.
What should the email ask for?
One thing. A few hours a week on a defined task, or permission to visit and see whether there is a fit. A request to “pick your brain” or “learn about your journey” does not give them anything to say yes to.
Should a parent send the email?
No. The student sends it, with a one-page résumé attached. A parent email tells the lab that the student is not the one who will show up.
When do you follow up?
Once, about a week later, on the same thread. If there is still no reply, write to a different person or apply to a program that already assigns a mentor. A third note to the same inbox does not create a spot.
Sources
- Texas Tech, Clark Scholars program details, for the cohort size and the stipend, and the application page for the February 15, 2027 deadline.
- Fermilab, PRISM, for the $500 weekly stipend and the Illinois high-school requirement.
- Stony Brook, Garcia Center, for the June 24, 2026 summer start.