Coefficient Giving — Technical AI Safety
Junior grantmaking on Coefficient Giving's Technical AI Safety team ($150M+ in grants in 2025). Fellows will co-lead grant investigations, scope new active grantmaking projects (incl. founding new orgs), and help shape the team's strategy and future RFPs.
Stream overview
- Fellows would be flexible junior grantmakers, who could do many potential activities, as listed below. Which activities in particular would be influenced by the team’s needs, the fellow’s individual strengths, e.g. how technical they are, how strategic their takes are, how professional / comfortably public-facing they seem…
- Potential activities:
- Grant investigations and write-ups, in collaboration with their mentor.
- Active grantmaking:
- Building infrastructure to support the TAIS field and to support founding new organisations
- I/N/T (importance, neglectedness, tractability) writeups on potential active grantmaking projects, including interviewing external experts.
- Conversations with potential grantees to pitch them on important neglected work
- Passive grantmaking
- Processing inbound proposals
- Writing a section of a future RFP, based on guidance from their mentor. (These usually involve time-intensive literature reviews, plus gathering takes from lots of external people.)
Mentors
Jake Mendel is a technical AI safety program officer at Coefficient Giving, where he makes grants for technical work. Previously, he worked as a research scientist at Apollo Research.
Lucas is a program operations associate on Coefficient Giving’s Technical AI Safety team. He previously worked as an associate consultant at Bain & Company and holds a Master in Public Policy from Harvard Kennedy School.
Ben Bateman is chief of staff for Technical AI Safety at Coefficient Giving. Previously, he was director of operations at Mirror Biology Dialogues Fund and spent six years at GiveWell.
Max Nadeau is a Program Officer on Coefficient Giving's Technical AI Safety team. Previously, he conducted research on machine-learning robustness and interpretability.
Mentorship style
Active mentorship, similar to a regular manager relationship, with at least one 1-1 per week and async guidance.
Fellows we are looking for
You can find a full list of criteria in this document.
In short, we are looking for a combination of:
- General skills
- Clear communication
- Critical thinking
- Good judgment
- Ownership and agency
- Pragmaticism
- Technical AI Safety knowledge/background, e.g. – this is slightly less important for generalist profiles
- Co-authorship of relevant research papers
- Relevant educational background
- Participation in previous TAIS mentorship programs
- Professional experience with a technical AI safety organization
Project selection
Fellows will work on a range of projects throughout the fellowship, and will work closely with their mentor to assign and execute on these projects. Fellows will have a comparable amount of freedom as a new grantmaker joining our team does.
The Winter 2026 cohort offers a wide range of research streams led by experts across AI alignment, interpretability, governance, and safety. Each stream provides its own research agenda, methodology, and mentorship focus.
