RL Research Scientist, Foundation Models & Self-Play

Lever, Inc.; Sunnyvale (CA); On-site; USD 150,000 - 450,000; Full time

Job summary

MBZUAI seeks a Research Scientist to advance reinforcement learning for foundation models. You will prototype novel RL methods, contribute to training infrastructure, and publish results through collaborations with researchers, data scientists, and engineers.

Based in Sunnyvale, the role emphasizes cross-disciplinary work to push emergent capabilities and robust deployment in real-world settings.

Qualifications

• MSc/MEng or PhD in ML, CS or related field.
• Minimum 3+ years hands‑on reinforcement learning experience.
• Strong Python development for research-grade code and scalable data pipelines.
• Experience applying novel RL algorithms to practical applications.
• Strong publication record or open-source contributions.

Responsibilities

• Develop novel research toward scalable self-play for foundation-model training.
• Initiate RL algorithmic approaches to drive emergent capabilities in Foundation Models.
• Full-stack engineering from data curation to production of models with well-documented code.
• Contribute to technical reports and research publications.
• Represent MBZUAI at conferences and engage with the open-source community.
• Contribute to large-scale RL training and inference frameworks.

Skills

Reinforcement learning

Python

Open-source

Publications

Education

MSc/MEng or PhD in ML/CS

Tools

PyTorch

Jax

Diffusion models

Job description

MBZUAI seeks a Research Scientist to advance reinforcement learning for foundation models. You will prototype novel RL methods, contribute to training infrastructure, and publish results through collaborations with researchers, data scientists, and engineers.

Based in Sunnyvale, the role emphasizes cross-disciplinary work to push emergent capabilities and robust deployment in real-world settings.