AI Research Scientist — Multi-Agent

About the role

Conduct fundamental research on multi-agent systems at DeepMind. You'll explore coordination, communication, and emergent behavior in populations of AI agents — with the goal of building systems that can collaborate effectively on complex real-world tasks.

Responsibilities

• Design and run experiments on multi-agent coordination and communication
• Develop novel algorithms for agent collaboration and task decomposition
• Publish research at top-tier venues (NeurIPS, ICML, ICLR)
• Collaborate with engineering teams to translate research into products
• Mentor junior researchers and interns

Requirements

• PhD in Machine Learning, Computer Science, or related field
• Published research in multi-agent systems, RL, or related areas
• Strong implementation skills in Python and JAX/PyTorch
• Ability to design clean experiments and draw rigorous conclusions
• Excellent communication and collaboration skills