Senior Reinforcement Learning Scientist

About the Role

We're building AI systems capable of autonomous mathematical discovery. As a Senior RL Scientist, you'll develop novel algorithms for theorem discovery and proof search, working at the intersection of reinforcement learning, mathematics, and language modeling.

Key Responsibilities

• Design RL algorithms for mathematical reasoning and long-horizon proof search
• Scale distributed training systems for complex mathematical domains
• Work with mathematicians and industry leaders to measure progress in AI mathematical discovery

Requirements

• PhD in ML/CS/Math or equivalent experience
• Deep expertise in RL theory and algorithms
• 3+ years with large-scale distributed RL training
• Track record of impactful ML publications
• Strong mathematical background and engineering skills

Bonus: Experience with theorem provers (Lean, Coq), RLHF, multi-agent systems, or publications at NeurIPS/ICML/ICLR.

Compensation & Logistics

Salary: Above market rate, equity and benefits
Location: San Francisco, CA
Visa: We sponsor for exceptional candidates

Join us in building AI that can make genuine mathematical discoveries. We combine rigorous research with the resources of a well-funded lab and the focus of a small team.

To apply: harry@principialabs.org with CV and research highlights.