USA remote
Member of Technical Staff, Physics Research
About this role
OVERVIEW Physical Superintelligence is a startup with roots at Google, NVIDIA, Harvard, Meta, MIT, Oxford, Johns Hopkins, Cambridge, and the Perimeter Institute building AI systems to discover new physics at scale. We are seeking fundamental physicists, theorists and experimentalists, to run research campaigns on open problems with AI systems, and to make the results verifiable. Our mission is to discover and commercialize transformative physics breakthroughs at scale with artificial superintelligence, safely, verifiably, and for broad public benefit.
The last century's golden age of physics gave us transistors, lasers, and nuclear energy. We believe artificial superintelligence will unlock the next one. We're creating the infrastructure to industrialize scientific discovery and usher in this new era. We have one product: new physics, at scale. ROLE AND RESPONSIBILITIES - Run research programs on open problems in your area: formulate, direct AI systems at them, judge what comes back, own the result.
- Make physics legible to machines: encode what counts as a correct answer, verify AI output against it, and stand behind results before they leave the building. - Go deep in one lane: quantum theory, many-body and condensed matter; statistical mechanics and complex systems; atomic, molecular and optical; high energy, particle and nuclear; gravity and cosmology; mathematical physics. - Write the manuscripts and own the path to publication, working with AI researchers on how agents reason about physics.
WHAT WE'RE LOOKING FOR - Three or more years of research experience, with publications in top-tier journals in one area of fundamental physics, and rigor with a certificate behind it. You derive results and check them; you know when a numerical hint is not a proof and when a signal is an artifact of the setup; and you reach for exact or SDP certificates, interval arithmetic, computer algebra, proof assistants, or large-scale numerical experiments with the artifact ruled out, when the problem calls for them.
- Three or more years of hands-on work at the physics-ML intersection. You have used machine learning or AI systems on a physics problem, trained or directed a model, and can read a modern ML paper, reason about what an agent is doing, and say how to evaluate its output. - Deep expertise in one area of theoretical or experimental physics: - Quantum theory, many-body and condensed matter. - Statistical mechanics and complex systems.