Research Scientist – AI Alignment

About the role

Resolution is seeking Research Scientists to advance AI alignment, focusing on scalable oversight and related research programs. You will work within a collaborative team, contributing both empirical and theoretical work to align artificial superintelligence.

Key responsibilities

• Translate and model alignment problems into concrete empirical or theoretical projects and execute them.
• Design and run experimental protocols, develop theoretical proofs, and contribute to autoformalization efforts.
• Write and present research outcomes as papers, blog posts, and talks.
• Communicate progress and challenges daily via Slack and in weekly team meetings.

Required profile

• Graduate degree (Ph.D.) or equivalent experience in ML, CS, mathematics, physics, statistics, philosophy, or related fields.
• Demonstrated research track record with strong technical writing (papers, preprints, etc.).
• Strong mathematical background, even for primarily empirical work.
• Motivation to work full‑time on AI superintelligence alignment.

Required skills

• Machine learning (including RLHF at frontier scale).
• Game theory, complexity theory, mechanism design, decision theory.
• Theoretical computer science and formal modeling.
• Bayesian statistics, information theory, optimization theory.

What we offer

• Salary range from $141,000 to $930,000 depending on level and location.
• 5 weeks paid vacation plus public holidays.
• Comprehensive health insurance (medical, dental, vision) and unlimited sick leave.
• Unconditional 4% 401(k) contribution.
• Visa sponsorship for relocation to Berkeley.