Staff Physicist for AI-Driven Causal Modeling

Causal; San Francisco (CA); On-site; USD 180,000 - 240,000; Full time

Job summary

Causal is building a Large Physics foundation Model to understand and predict physical systems, starting with weather and governed by physical law.

We seek domain experts to bring physics to bear on the model, develop evaluations of physical coherence, and advise on the physics of the systems we model.

You will collaborate with ML researchers and evaluation teams, exploring where physics-informed inductive biases help or hinder and how generalization occurs.

Qualifications

• Deep expertise in physics with focus on fluid dynamics or thermodynamics.
• Experience with numerical simulation of physical systems (CFD) and its trade-offs.
• Interest in combining machine learning with physical modeling.
• Ability to collaborate with ML researchers and translate physics into requirements.
• Rigorous, evidence-driven approach to evaluating model quality.

Responsibilities

• Bring physical principles to bear on the model and assess conservation laws.
• Develop evaluations for physical coherence beyond statistics.
• Advise on physics of modeled systems, from fluids to thermodynamics.
• Investigate LPM generalization across physical domains and breakdowns.
• Collaborate with model, evaluation, and interpretability teams to guide research.

Skills

Fluid dynamics

Thermodynamics

Computational physics

CFD familiarity

ML-physics intersection

Collaboration with ML researchers

Evidence-driven evaluation

Education

PhD or equivalent research experience

Job description

Causal is building a Large Physics foundation Model to understand and predict physical systems, starting with weather and governed by physical law.

We seek domain experts to bring physics to bear on the model, develop evaluations of physical coherence, and advise on the physics of the systems we model.

You will collaborate with ML researchers and evaluation teams, exploring where physics-informed inductive biases help or hinder and how generalization occurs.