Research Scientist - Reinforcement Learning

What you'll do

• Develop self-play and other RL methods for foundation-model training, agentic tasks, and learning from environments.
• Engineer end-to-end research and production workflows, contribute to large-scale RL training and inference frameworks, and release documented, tested code.
• Publish and present research, engage with the open-source community, and collaborate with internal and external researchers.

What they ask for

• MSc/MEng or PhD in Machine Learning, Computer Science, or a related field, or equivalent experience.
• 3+ years of hands-on reinforcement learning experience.
• Strong Python skills, including research-grade code and scalable data pipelines; experience implementing mathematical concepts as reliable, documented code.
• Experience applying novel RL algorithms to practical applications and contributing to academic or open-source research through publications, GitHub contributions, or presentations.

Nice to have

• Expertise in deep learning frameworks such as PyTorch or Jax; experience with large-scale model training or current RL+LLM training libraries.
• Experience with self-play or Diffusion Models in RL; strong publication record in leading AI/RL venues.

Pay

USD 150,000–450,000 per-year-salary (base salary); $150,000 – $450,000/year (base compensation)

Tools

Python, PyTorch, Jax, LLMs, Diffusion Models

AI focus

Develop novel reinforcement learning methods, massive-scale self-play, and agentic capabilities for foundation models, including learning from experience and environments.