Research Scientist- Reinforcement Learning (Open-Endedness)

Job Description

Member Technical Staff- Reinforcement Learning & Open-Ended Learning. Full-time, permanent opportunity in San Francisco.

We're representing an early-stage applied research lab building AI capable of open-ended learning, systems that keep getting better by discovering their own goals rather than optimizing ones we hand them

What you'll do:

• Develop RL methods for agents that discover useful objectives, tasks, and curricula without relying entirely on human-specified rewards
• Design systems for open-ended learning including unsupervised / automated environment design, asymmetric self-play, and intrinsic motivation
• Build training loops where agents learn from interaction, exploration, novelty, competence progress, self-generated challenges, or other nonstandard reward signals
• Investigate how agents can avoid collapsing into trivial, degenerate, or easily exploitable objectives
• Own and develop a research agenda end to end — from identifying promising directions, to running experiments, to communicating results

What we're looking for

• 5+ years in reinforcement learning research (PhD and academic years count).
• Strong RL fundamentals, with exposure to one or more of: open-endedness, quality-diversity methods, intrinsic motivation, self-play, multi-agent RL, or goal-conditioned RL.
• Proficiency in Python and modern ML frameworks (PyTorch and/or JAX)
• A track record of research output and/or shipping research-grade code.
• High agency and comfort operating in an early-stage, fast-moving research environment.

Nice to have

• Experience with LLM post-training and/or coding agents.
• Familiarity with population-based training, POET, or related open-ended / evolutionary approaches
• Publications at top venues (NeurIPS, ICML, ICLR, GECCO, RLC, or similar).