Full Program Details
Much of the current AI boom runs on NVIDIA hardware, and many of the ideas behind that hardware started in university PhD labs. NVIDIA backs that early research directly through its Graduate Fellowship Program. The 2027-2028 cycle is now open, and it closes on 30 October 2026.
The fellowship gives selected PhD students up to $60,000 for one academic year, paid through their university to cover stipend, tuition and mandatory fees. It also brings each fellow into NVIDIA Research for a summer internship before the award year starts.
Up to $60,000 toward your PhD stipend and fees, plus an in-person research internship at NVIDIA. International PhD students can apply.
Details at a glance:
- Host: NVIDIA Research
- Award: up to $60,000 for the 2027-2028 academic year, paid to your university
- Level: PhD students who have finished at least their first year
- Fields: computer science, computer engineering, system architecture, electrical engineering or a related area
- Applications opened: 30 September 2026
- Deadline: 30 October 2026, including recommendation letters
- Winners notified: 15 January 2027
What the Fellowship Covers
- Up to $60,000 toward your stipend, tuition and mandatory fees for the nine-month award year
- The money goes to your university, not to you directly, and it cannot exceed your actual stipend, tuition and fees
- A required in-person summer internship at an NVIDIA research office before the award year begins
NVIDIA notes that award amounts may vary by country or location, so the full $60,000 is a ceiling, not a promise.
Who Can Apply
- Full-time PhD students who have completed their first year of PhD study. NVIDIA says it considers exceptional cases, but applicants without PhD experience are very unlikely to win.
- Students actively working on thesis research in computing or a related field
- Students who will stay enrolled full time for the whole award year and do not plan to graduate before it ends
- International students are welcome. NVIDIA does not limit the number of applicants per university or per research group.
- NVIDIA employees and their immediate family members cannot apply
You must also be able to do the internship in person at one of NVIDIA's eligible research offices: the United States, Canada, Germany, Sweden, Finland, Switzerland, Israel, Taiwan or Singapore. If you study in Africa, plan for the visa and travel that this requires.
Research Areas NVIDIA Is Looking For
NVIDIA lists a wide spread of topics, including generative AI, language models and speech, computer architecture, computer graphics, robotics, autonomous vehicles, parallel programming, security and privacy, and quantum computing.
How to Apply
- Confirm with your advisor that you can take a summer internship at an eligible NVIDIA office before the award year.
- Write a research proposal of up to two pages, plus a bibliography that does not count toward the limit. Put your name at the top.
- Update your resume or CV with your contact details and publications.
- Ask two or three people for recommendation letters. At least two are required, and one must come from your thesis advisor or professor.
- Submit the proposal, CV, letters and internship confirmation through the portal linked on NVIDIA's fellowship page before 30 October 2026.
How NVIDIA Picks Winners
- Student quality: recommendation letters, GPA and achievements
- Research quality: results so far, your proposal and your publication record
- Relevance to NVIDIA: how closely your work connects to NVIDIA's research areas and could influence future GPUs
GizPulse Verdict
Who this is for: Second-year and later PhD students in AI, graphics, systems or hardware whose work could plausibly run on, or change, the next generation of GPUs.
Who should pause: Master's students, first-year PhD students with no research results yet, and anyone who cannot travel for an in-person summer internship in one of the listed countries.
One tip to stand out: NVIDIA scores relevance to its own research directly. In your two pages, name the exact NVIDIA problem your work helps with, such as faster training, better rendering or safer robots, and show one result you already have.
Read the full rules and submit your application on the official NVIDIA Graduate Fellowship page.
