Director, Physical AI for Autonomous Discovery

What you'll do

• Define and execute the Physical AI strategy and roadmap for autonomous laboratory experimentation.
• Lead a high-performing team to develop models that perceive, reason, and act within laboratory environments.
• Build AI models to understand instruments, samples, reagents, robotic systems, and overall laboratory states.
• Develop adaptive experimental optimization capabilities using active learning, Bayesian optimization, and reinforcement learning.
• Create systems for physical error detection, root-cause assessment, and automated workflow replanning.
• Design agentic and multi-agent systems to coordinate experiment planning, execution, and data interpretation.
• Partner with engineering and discovery science teams to deploy reliable closed-loop learning in real laboratories.
• Embed uncertainty calibration, safety constraints, and governance into autonomous decision-making processes.

What we're looking for

• PhD in artificial intelligence, machine learning, robotics, computer science, engineering, computational biology, or a related quantitative field.
• Demonstrated experience developing AI systems that operate in physical or scientific environments.
• Deep expertise in world models, reinforcement learning, Bayesian optimization, causal inference, multimodal AI, robotics, or scientific machine learning.
• Strong software-development skills including Python and modern machine-learning frameworks.
• Ability to translate scientific and operational requirements into robust, scalable AI models and communicate across multidisciplinary teams.
• Strong collaboration skills to work in a matrix environment.
• Experience with self-driving laboratories, laboratory automation, or closed-loop scientific experimentation (preferred).
• Experience with vision-language-action models, digital twins, machine vision, robotic control, or multi-agent systems (preferred).