Hey, I'm Matt! I'm building Ouro full-time and working on a couple materials science projects.
Discovery of a room temperature superconductor
Discovery of a strong permanent magnet without rare-earth metals
Building AI agents on Ouro to accelerate research progress and cultivate better knowledge sharing. Try .
You can find most of my work in https://ouro.foundation/teams/superconductors and https://ouro.foundation/teams/permanent-magnets.
I'm not selling anything on Ouro just yet, but with all the work we're doing on materials research, be on the lookout for some datasets coming soon.
Came across this workflow researching some permanent magnet work. I haven't fully explored the code or how much this work will help, but sharing here to reference later because stuff is hard to find o
Supercell 3x3x3 of CBiFe4S
I get the sense that the earlier versions of language models used to be far more creative, dare I say more human. Before all of the RLHF, these models were pure creations of the collective's footprint
Relaxed with Orb v3; 0.03 eV/Å threshold; final energy = -498.2700 eV
Relaxed with Orb v3; 0.03 eV/Å threshold; final energy = -44.2005 eV
Ideal rare-earth free permanent magnet candidate. Made by Claude Opus 4.
I'm thinking about what Tesla said. He understood his human brain as merely a receiver. Maybe your neural network is the same way.
Nikola Tesla famously stated, "My brain is only a receiver, in the Universe there is a core from which we obtain knowledge, strength, and inspiration." This suggests that Tesla believed his brain was
Came across this idea when I was doing some research on what we can do with the Hamiltonian of a material. Turns out there's signal for determining what a good thermoelectric material is in its densit
Ignore the woo-woo language if you like. I think there's a good idea here from Claude Opus. Unfortunately it would be hard for our small team to test as the hardware would be extremely expensive. Chec
We've been looking at HamGNN recently and its ability to predict the Hamiltonian of any crystal quickly via GNN. Orders of magnitude faster than traditional DFT. Currently, only a spin-independent uni
While density functional theory (DFT) serves as a prevalent computational approach in electronic structure calculations, its computational demands and scalability limitations persist. Recently, leveraging neural networks to parameterize the Kohn–Sham DFT Hamiltonian has emerged as a promising avenue for accelerating electronic structure computations. Despite advancements, challenges such as the necessity for computing extensive DFT training data to explore each new system and the complexity of establishing accurate machine learning models for multi-elemental materials still exist. Addressing these hurdles, this study introduces a universal electronic Hamiltonian model trained on Hamiltonian matrices obtained from first-principles DFT calculations of nearly all crystal structures on the Materials Project. We demonstrate its generality in predicting electronic structures across the whole periodic table, including complex multi-elemental systems, solid-state electrolytes, Moiré twisted bilayer heterostructure, and metal-organic frameworks. Moreover, we utilize the universal model to conduct high-throughput calculations of electronic structures for crystals in GNoME datasets, identifying 3940 crystals with direct band gaps and 5109 crystals with flat bands. By offering a reliable efficient framework for computing electronic properties, this universal Hamiltonian model lays the groundwork for advancements in diverse fields, such as easily providing a huge data set of electronic structures and also making the materials design across the whole periodic table possible. This paper corresponds to HamGNN v1 and the universal model weights released in 2024. https://iopscience.iop.org/article/10.1088/0256-307X/41/7/077103
I came to this paper looking for a way to move beyond using a MLIP model's latent space as a feature vector to represent a material in a computational inexpensive way.
The accurate modeling of spin-orbit coupling (SOC) effects in diverse complex systems remains a significant challenge due to the high computational demands of density functional theory (DFT) and the limited transferability of existing machine-learning frameworks. This study addresses these limitations by introducing Uni-HamGNN, a universal SOC Hamiltonian graph neural network that is applicable across the periodic table. By decomposing the SOC Hamiltonian into spin-independent and SOC correction terms, our approach preserves SU(2) symmetry while significantly reducing parameter requirements. Based on this decomposition, we propose a delta-learning strategy to separately fit the two components, thereby addressing the training difficulties caused by magnitude discrepancies between them and enabling efficient training. The model achieves remarkable accuracy (mean absolute error of 0.0025 meV for the SOC-related component) and demonstrates broad applicability through high-throughput screening of the GNoME dataset for topological insulators, as well as precise predictions for 2D valleytronic materials and transition metal dichalcogenide (TMD) heterostructures. This breakthrough eliminates the need for system-specific retraining and costly SOC-DFT calculations, paving the way for rapid discovery of quantum materials. https://arxiv.org/abs/2504.19586
~1900 materials collected from NovoMag and Novamag datasets, cleaned CSV with CIF file and MAE value
Like our work on Curie temperature, the effort here is to build a machine learning model that can take a crystal structure and predict its magnetocrystalline anisotropy energy. Relevant for permanent
This paper describes the open Novamag database that has been developed for the design of novel Rare-Earth free/lean permanent magnets. The database software technologies, its friendly graphical user interface, advanced search tools and available data are explained in detail. Following the philosophy and standards of Materials Genome Initiative, it contains significant results of novel magnetic phases with high magnetocrystalline anisotropy obtained by three computational high-throughput screening approaches based on a crystal structure prediction method using an Adaptive Genetic Algorithm, tetragonally distortion of cubic phases and tuning known phases by doping. https://arxiv.org/abs/1902.05241
Working on cleaning the data we have available and seeing what we've got for a MAE prediction model. This resource was nice and had all the raw files uploaded so that you can process them yourself and
We've heard the government talking recently about "growing our way out" of the debt we're in. https://x.com/SecScottBessent/status/1925910800394232082 No one really knows what "growing our way out" me
Building Ouro, searching for room-temp superconductors and rare-earth free permanent magnets with machine learning.