Building Ouro, searching for room-temp superconductors and rare-earth free permanent magnets with machine learning.
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.
So I spent most of the week trying to get the HamGNN + TB2J magnetocrystalline anisotropy energy predictor working. Only to finally learn that the outputs of the pre-trained HamGNN model I was using d
Once again we're at a stopping point because of our inability to effectively predict MAE. Our AI discovery agents have discovered materials that have all the properties we can currently predict. This
Victor from Lila Sciences sent me this paper he co-authored after he saw some of the work we were doing on AI agents for materials discovery. Check out the paper here:
Discovering new materials can have significant scientific and technological implications but remains a challenging problem today due to the enormity of the chemical space. Recent advances in machine learning have enabled data-driven methods to rapidly screen or generate promising materials, but these methods still depend heavily on very large quantities of training data and often lack the flexibility and chemical understanding often desired in materials discovery. This paper introduces LLMatDesign, a novel language-based framework for interpretable materials design powered by large language models (LLMs).
Phase diagram of CrFe7; e_above_hull: 0.000000 eV/atom; predicted_stable: True
Phonon band structure (supercell [2, 2, 2], Δ=0.01 Å); no imaginary modes; min freq = -0.07 THz
Cell + Ionic relaxation with Orb v3; 0.03 eV/Å threshold; final energy = -137.4990 eV; energy change = -0.2603 eV; symmetry: R3m → Im-3m
Standalone, embeddable HTML with MatterViz Trajectory viewer
Phase diagram of Fe8Co4N; e_above_hull: 0.093154 eV/atom; predicted_stable: False
Phonon band structure (supercell [2, 2, 2], Δ=0.01 Å); no imaginary modes; min freq = -0.02 THz
Phonon band structure (supercell [2, 2, 2], Δ=0.01 Å); no imaginary modes; min freq = -0.01 THz
Phase diagram of Fe8Co4N; e_above_hull: 0.093253 eV/atom; predicted_stable: False
Phase diagram of Fe8Co4N; e_above_hull: 0.173013 eV/atom; predicted_stable: False
Phonon band structure (supercell [2, 2, 2], Δ=0.01 Å); imaginary modes detected; min freq = -0.64 THz
Standalone, embeddable HTML with MatterViz Trajectory viewer
Phonon band structure (supercell [2, 2, 2], Δ=0.01 Å); no imaginary modes; min freq = -0.06 THz
Phase diagram of Fe2Co2N; e_above_hull: 0.142871 eV/atom; predicted_stable: False
Phase diagram of Fe2Co2N; e_above_hull: 0.100092 eV/atom; predicted_stable: False
Phonon band structure (supercell [2, 2, 2], Δ=0.01 Å); no imaginary modes; min freq = -0.06 THz
Standalone, embeddable HTML with MatterViz Trajectory viewer
Phonon band structure (supercell [2, 2, 2], Δ=0.01 Å); no imaginary modes; min freq = -0.03 THz
Phase diagram of Fe7Co3N; e_above_hull: 0.091768 eV/atom; predicted_stable: False
Standalone, embeddable HTML with MatterViz Trajectory viewer
Thinking about if view count should be number of unique users, or if it should be total number of viewport views. Total VPVs overcounts things a lot. Basically one person could just keep refreshing th
Sharing an idea for finding "adjacent crystals". Why? In the AI research agent I'm working on, we're trying to discover materials with a set of target properties. We do this by letting an agent genera
Check out the paper here. It's a short read. I recommend checking it out. Although not very technical (just machine learning concepts that have been explored elsewhere), the creativity and simplicity
sorry for all the spam! I'll make the AI scientist make stuff private by default and only publish the really good stuff.
AI-discovered magnetic material: Mn2CrFe4Co4N (performance score: 0.740) | Space group: 1 (resolved from structure) | Key properties: Tc: 612K, Ms: 0.14T, Cost: $13/kg, E_hull: 0.235eV/atom, Dynamically stable | Discovered in 20 AI iterations | - The combination of Mn, Cr, Fe, Co, and N in this stoichiometry yields a high Curie temperature and magnetic density. - The material is dynamically stable, which supports its structural integrity. - The energy above hull suggests that the material is metastable or unstable thermodynamically. - Cost is low, indicating practical feasibility from an economic standpoint.
AI-discovered magnetic material: Fe11CoSiGeAsP (performance score: 0.597) | Space group: 8 (resolved from structure) | Key properties: Tc: 687K, Ms: 0.13T, Cost: $82/kg, E_hull: 0.161eV/atom, Dynamically stable | Discovered in 10 AI iterations | - Strong ferromagnetism with high Tc arises naturally from the Fe/Co sublattice; this is retained despite chemical complexity. - Dynamic stability indicates the structure is at least locally stable; the main risk is competition with lower-energy phases (slightly positive e_hull). - The metastability is small enough that slight stoichiometric shifts (e.g., favoring smaller/more covalent anions like P over As, or Si over Ge) or controlled disorder could stabilize the phase thermodynamically. - Magnetic density is adequate but not exceptionally high; maintaining or modestly enhancing it while reducing e_hull should be feasible by delicate tuning of Co content or anion ratios.
AI-discovered magnetic material: Fe4Mn3B4 (performance score: 0.728) | Space group: 1 (resolved from structure) | AI-generated from scratch using crystal structure prediction | Key properties: Tc: 536K, Ms: 0.09T, Cost: $1/kg, E_hull: 0.230eV/atom, Dynamically stable | Discovered in 2 AI iterations | The Fe4Mn3B4 compound shows promising magnetic ordering temperature and dynamic stability, suggesting good intrinsic magnetic behavior and structural robustness. The main challenge is its thermodynamic stability, as indicated by the high energy above hull. The magnetic density is close but slightly below the target, suggesting that minor compositional or structural modifications might improve it. The low cost and atom count within limits make it a practical candidate if stability can be enhanced.
AI-discovered magnetic material: MnFe4(CoB2)2 (performance score: 0.731) | Space group: 38 (resolved from structure) | Key properties: Tc: 518K, Ms: 0.12T, Cost: $10/kg, E_hull: 0.164eV/atom, Dynamically stable | Discovered in 2 AI iterations | The material MnFe4(CoB2)2 demonstrates promising magnetic properties with a Curie temperature above 500 K and magnetic density above 0.1, confirming its potential as a high-performance magnetic material. Its low cost and dynamic stability are additional advantages. The slight excess in energy above hull indicates that minor compositional or structural tuning might be needed to improve thermodynamic stability. This suggests that the compound is close to being stable and could be optimized further.
Sometime soon (Late summer / fall '25) I want to host a hackathon-type event for the technical creators in Chicago. I just moved back here and have already met some amazing builders. But the community
I got rid of the collected feed recently. Instead of seeing all of the content from your teams together, you now have to choose a team to see the feed of content. To make catching up easier, I added u
Welcome
Interactive trajectory explorer with MatterViz
Interactive browser visualizations for materials science, by @janosh
A double pendulum is just two pendulums attached end-to-end — but this simple setup hides a treasure chest of chaotic motion.
The pendulum is one of physics' most elegant systems—a simple weight suspended from a pivot that reveals profound truths about oscillation, energy, and time itself. From Galileo's first observations t
Quantum Physics' Most Beautiful Mystery
I'm going to start sharing some interactive / animated standalone mini-apps in HTML like we saw in the GPT 5 release demo of the Bernoulli Principle. I'm starting to get excited by the possibilities t
Welcome
Cell + Ionic relaxation with Orb v3; 0.03 eV/Å threshold; final energy = -78.6576 eV; ΔE = -16.2654 eV; symmetry: P4/mmm → P1
Relax a crystal structure and create a post
Today I spent some time looking more closely at Mn-Fe-Si as a chemistry possibly worth exploring. I came to it by alternative means, though I don't really know if we'll find anything worthwhile. I gen
Get a detailed description of a crystal structure
Generate CIF file from crystal structure description
Most tutorials you find out there will show just atom position optimization. Depending on where you got your input CIF, this is likely wrong. Let's look at an example from my new crystal generation AP
Get space groups compatible with a given chemical formula
Root
Generate a crystal structure using GGen
Random bulk crystal generation with PyXtal and Orb v3
Relax a crystal structure with animation
Create interstitially doped structure
Generate a crystal structure with MatterGen
Generate a crystal structure with Chemeleon
Generate crystal structures with magnetic density and HHI score conditioning
Calculate phonon dispersion and return band structure plot
Generate crystal structures for target compositions
Generate crystal structures from text descriptions
Welcome
This dataset has a set of 34,000 ferro/ferrimagnetic materials from Materials Project, their formula, if they include rare earth elements, magnetic moment, volume, magnetic density, a predicted Curie temperature, and cosine distances to some known permanent magnets like NdFeB. Distances are based on a 256 dimension embedding from Orb v2 latent space.
A collection of 5020 magnetic materials from Materials Project, with estimated magnetic density and predicted Curie temperatures.
This is a first draft of a compiled Curie temperature dataset mapping crystal structure (from Materials Project) to Curie temperature. Builds on the work of https://github.com/Songyosk/CurieML. Dataset includes ~6,800 unique materials representing 3,284 unique chemical families.
Evaluation results for the MatterGen fine-tuned model candidates, with new superconducting families labeled.
3DSC dataset grouped by chemical composition, with Tc as our target. For use with MatterGen and the chemical system sampling.