Summer research fellowships for researchers and builders who want to work on open and challenging problems on self-improving AI systems.
The fellowship
The NeoSigma Research Fellowship is a 10-14 week, in-person, paid program for exceptional researchers and builders to work on open, real-world AI challenges.
Fellows collaborate closely with our research team on problems at the frontier of self-improving AI systems. Projects are intentionally open-ended—we work with each fellow to define a direction that aligns with their strengths, interests, and long-term goals.
We strongly encourage fellows to contribute to the broader research community—through publications, technical blogs, and open research artifacts—helping share learnings and advance the field.
Who should apply
Folks who combine a strong research mindset with a bias for action. If you've taken a research idea from scratch to something real end-to-end, we'd love to hear from you. We welcome applications from diverse backgrounds including:
Researchers & PhD Students
Graduate students or early researchers in ML, systems, or adjacent fields who want to work on unsolved frontier problems outside a traditional academic setting.
Strong Engineers
Engineers with a track record of building real systems who want to go deeper into research — you care about understanding why things work, not just making them work.
Self-Taught Builders
You've learned by building. You've shipped things end-to-end and have something real to show for it — a project, a paper, a system, or a tool others rely on.
The experience
Real, open research problems
You'll be embedded in our core research team from day one, working on real-world open-ended research problems.
Direct, involved mentorship
Work side-by-side with our research team and get direct involved mentorship, feedback and brainstorming sessions.
Competitive paid stipend
All fellows receive a competitive paid stipend and other office perks.
Research output
Fellows directly contribute to research outputs — papers, technical blogs, and artifacts shared openly with the broader research community.
Apply
If you're excited about working on self-improving AI systems, we'd love to hear from you.
FAQ