People
The people behind the work.
Phanguard is a small founding team working in the open toward careful ML-safety research — and we're growing.
Core team
The founders
The two founders leading Phanguard's research agenda and the program.
Alexander leads Phanguard's research. He works on mechanistic interpretability and is currently co-authoring an in-progress paper on reasoning via sparse autoencoders. He has presented work at the NEMI workshop and reviewed for the EvalEval workshop at ACL 2026. He cares about open, careful research and about helping students get into the field.
Mentors
Mentorship
Every student in our first cohort is paired with a volunteer mentor for technical guidance and feedback. We're recruiting our founding group of mentors now — experienced researchers and practitioners who can give a few hours a week to one or two students.
Interested in mentoring?
We'd love to hear from you. The commitment is flexible, and you'll be paired with students whose work fits your background.
Become a mentor
We're always looking for experienced researchers to guide our students. The time commitment is flexible — a few hours a week.
Mentor with us →Join the program
The research program is free and open to high-school students aged 13–18. We're accepting 12 students for our first cohort.
Apply as a student →