Syntelesis Fellowship
A program for early-career researchers committed to advancing the science of multi-agent AI safety.
The Fellowship
The Syntelesis Fellowship supports exceptional researchers and engineers at the early stages of their careers who are dedicated to understanding and securing multi-agent AI systems. We provide Fellows with mentorship from leading researchers, access to our network of collaborators and funders, and the resources to pursue foundational work in a field still being defined.
The Program
We believe that early access to the right people, problems, and platforms can shape a researcher's trajectory and the trajectory of an emerging field.
The Fellowship offers:
-
Direct mentorship from Syntelesis Lab researchers and senior advisors
-
Access to our network of collaborators across AI safety, governance, and complex systems
-
Invitations to workshops, reading groups, and research convenings
-
Opportunities to present and publish work with Lab support
-
Guidance on research direction, career development, and funding pathways
The program is flexible and designed around each Fellow's research goals. Our focus is on enabling high-potential individuals to do their best work in multi-agent AI safety, with the support and community to sustain it.
What We Look For
We seek individuals drawn to hard, unsolved problems at the intersection of AI systems, collective behavior, and safety.
Research Focus: We prioritize applicants working on topics aligned with our mission, including but not limited to:
-
Multi-agent interaction dynamics and emergent behavior
-
Evaluation and benchmarking of collective AI systems
-
Contagion, propagation, and network effects in agent populations
-
Security vulnerabilities in multi-agent environments
-
Governance, policy, and accountability for multi-agent AI
Intellectual Depth: Applicants should demonstrate strong technical foundations and a capacity for rigorous, independent research. Publications, notable projects, or meaningful contributions to the field are valued.
Clarity of Purpose: We look for researchers who understand why multi-agent safety matters and can articulate what they want to work on and why.
Collaborative Spirit: The problems we address require insight from many disciplines. We seek Fellows who engage openly, share generously, and build with others.