AI Capabilities
AI that generates, screens, and hunts for new materials
Three research papers and one autonomous search system — here's what each one does.
149,419
Candidates generated
8,800
Flagged likely metallic
3
Published research methods
Framework Generation
A fine-tuned machine-learning model searches thousands of known framework structures for new magnetic variants.
Metallic Screening
A neural network ranks candidates before anyone runs a costly quantum simulation.
Agentic Search
Autonomous agents propose, test, and refine candidates on their own.
Qubit Characterization
A convolutional neural network reads spectroscopy images to measure defects in quantum chips.
Paper 1
Generating magnetic metal-organic frameworks
Metal-organic frameworks could replace rare-earth magnets, but their large unit cells make first-principles simulation too slow to search at scale. We fine-tuned a machine-learned interatomic potential (CHGNet) on 15,000+ first-principles computations from our Database, then used it to drive a site-substitution search across structural prototypes from the Quantum Metal-Organic Framework database — generating novel, highly magnetic candidates far faster than density functional theory alone.
15,000+
Density functional theory calculations used to fine-tune CHGNet
149,419
Novel magnetic metal-organic framework candidates generated
12,500
Density functional theory band gaps used to train the SchNet neural network
8,800
(5.9%)
Candidates flagged as likely metallic
Paper 2
Screening for metallic character
A magnetic metal-organic framework is only useful for spintronics if it also conducts. Generating candidates is one problem; knowing which ones are metallic is another. We screened all 149,419 candidates from the companion generation study with a SchNet neural network trained on 12,500 density functional theory band gaps, flagging 8,800 (5.9%) as likely metallic — a prioritization tool that narrows the field before anyone runs an expensive density functional theory calculation.
Agentic AI
An autonomous system that searches for materials on its own
We run agentic AI — autonomous systems that explore, act, and learn on their own — to search for new materials. Point it at a target property, and a swarm of AI agents proposes, tests, and refines candidates on its own. Today that loop is focused on finding a rare-earth-free permanent magnet, but the same system adapts to whatever material problem you bring us.
Every candidate still has to earn its place — validated against real physics before it's ever considered real.
Paper 3
Characterizing qubit defects with machine learning
Superconducting qubits lose coherence to two-level system defects buried in their building blocks, and finding them one by one with traditional fitting is slow — especially across a chip with hundreds of qubits. We trained a ResNet-based convolutional neural network on simulated two-tone spectroscopy images to read off all four defect parameters — frequency, coupling strength, and both decay times — directly from the image, outperforming standard perturbation-theory fitting and accelerating the loop of designing better quantum chips.
4
Defect parameters predicted per scan — frequency, coupling, and both decay times
Deep neural network
ResNet-based, trained on simulated two-tone spectroscopy images
Partner With Us
Let's make discoveries together
Want our database, models or candidates, or a joint discovery project? We reply within one business day.