Periodic Labs builds AI systems and autonomous laboratory infrastructure designed to accelerate scientific discovery in the physical sciences. The company combines frontier AI models with physical experimentation, deploying AI agents that can form hypotheses, design experiments, and learn from real-world results. Its autonomous laboratories generate experimental data at scale, closing the feedback loop between computational models and physical reality.
The technical foundation spans physics, chemistry, and machine learning. Teams of physicists, chemists, and ML researchers work together on problems in materials discovery, including superconductors and semiconductor engineering. The company operates weekly teaching sessions where physicists instruct AI systems on quantum mechanics reasoning, while ML researchers develop their understanding of physics fundamentals. This bidirectional knowledge transfer reflects the belief that breakthrough discoveries require both sophisticated AI and grounding in experimental reality.
Periodic Labs serves customers in space, defense, and semiconductor sectors. Recent applications include helping semiconductor manufacturers address heat dissipation challenges. The company's approach rests on the premise that frontier AI models trained on internet text require real experimental feedback to reason effectively about physical systems, and that autonomous laboratories can provide that feedback at scale.
