MEDIUM confidence
Researched 2026-08-04 20:54 · profile: researcher
Programme Thesis
The Seed STEM Fellows Program funds a 6-month, on-site research collaboration with ByteDance's Seed AI team, aiming to accelerate scientific discovery by embedding top STEM researchers with AI/ML expertise into frontier AI research. It exists to bridge domain science and foundation-model development, offering two tracks (Scientific Advisor and Ph.D. Intern) for researchers to work on real-world challenges.
Selection Criteria
- Hold a Ph.D. in STEM, be currently enrolled in one, or demonstrate equivalent research expertise.
- Deep academic expertise and thorough understanding of one or more STEM disciplines.
- Strong interest in accelerating scientific discovery with AI, belief in its potential, and ability to assess AI's value and limitations in their field.
- Extensive use of AI as a productivity tool in day-to-day research, with proficiency in programming.
- For Ph.D. Intern track: must be a current Ph.D. student.
- For Scientific Advisor track: must be affiliated with an academic institution or be a senior industry expert.
- Application materials: a brief statement (1-2 key challenges in field + how AI is used), CV with 1-2 representative publications.
- On-site collaboration in Beijing (Haidian District) is required.
Past Winners / Cohort Profiles
The page does not list past winners, but the program targets '100 outstanding researchers' from diverse STEM fields. The two tracks suggest past cohorts include senior academics/industry experts (Scientific Advisors) and current Ph.D. students (Ph.D. Interns). No named examples are available from the page.
Ideal Candidate Fingerprint
The ideal applicant is a current Ph.D. student or senior researcher with deep domain expertise in a fundamental science (e.g., biology, chemistry, physics) and strong AI/ML skills, evidenced by publications and daily use of AI tools. They are excited to apply foundation models to accelerate discovery in their field and are willing to relocate to Beijing for 6 months.
Recommended Framing
For Eniola, the strongest angle is to apply as a Scientific Advisor (given his independent research profile and M.Sc. enrollment, not a Ph.D. student) and lead with TOPOLOGIX, his protein language model (ESM-2) approach to drug-resistance prediction, which directly matches the program's mission of using AI to accelerate scientific discovery. He should emphasize his hands-on use of AI (ESM-2, Random Forest, Bayesian methods) and his ability to assess AI's limitations (as shown in his hERG topology replication study), while framing his independent research as equivalent to Ph.D.-level expertise. However, he must address the on-site requirement and his current M.Sc. enrollment status.
Watch Out
Eligibility concern: The program explicitly targets current Ph.D. students (Ph.D. Intern) or senior experts (Scientific Advisor). Eniola is not a current Ph.D. student and may not yet qualify as a 'senior expert' (though his independent research and publications could argue for equivalent expertise). Also, the on-site requirement in Beijing may conflict with his M.Sc. enrollment in Germany and his employment. The program does not provide direct funding or equity, which may not meet his funding needs.