Stefan Bringuier

Multi-stage-Bayesian-optimisation. Jupyter Notebook

tt-atom. Meta's UMA and Orbital Materials' Orb-v3/OrbMol interatomic potential models, running on Tenstorrent hardware.

Unlimited-OCR. Unlimited OCR Works: Welcome the Era of One-shot Long-horizon Parsing.

quests. Quick Uncertainty and Entropy via STructural Similarity

headroom. Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.

trackio. A lightweight, local-first, and free experiment tracking library from Hugging Face 🤗

betapy. New and improved python project for projected force constants and their visualization

UQ-MLIP. Uncertainty quantification (UQ) for Machine-Learning Interatomic Potentials (MLIPs)

Math-To-Manim. Create Epic Math and Physics Animations & Study Notes From Text and Images.

le-wm. Official code base for LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels

codegraph. Pre-indexed code knowledge graph, auto syncs on code changes, for Claude Code, Codex, Gemini, Cursor, OpenCode, AntiGravity, Kiro, and Hermes Agent — fewer tokens, fewer tool calls, 100% local

academic-research-skills. Academic Research Skills for Claude Code: research → write → review → revise → finalize

screenpipe. YC (S26) | Record your screen 24/7 and plug into your agents. Local, private, secure. Connect to OpenClaw, Hermes agent and 100+ apps

amorphouspy. Workflows for computing glass properties at the atomic level

marker. Convert PDF to markdown + JSON quickly with high accuracy

HamGNN. An E(3) equivariant Graph Neural Network for predicting electronic Hamiltonian matrix

xxh. 🚀 Bring your favorite shell wherever you go through the ssh. Xonsh shell, fish, zsh, osquery and so on.

zellij. A terminal workspace with batteries included

andrej-karpathy-skills. A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.

claude-mem. Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More

graphify. Turn any codebase, with its docs, SQL schemas, configs, and PDFs, into a queryable knowledge graph. A /graphify skill for Claude Code, Cursor, Codex, and Gemini CLI: local deterministic AST parsing, every edge explained, no vector store.

skills. Public repository for Agent Skills

superpowers. An agentic skills framework & software development methodology that works.

pi. AI agent toolkit: unified LLM API, agent loop, TUI, coding agent CLI

mempalace. The best-benchmarked open-source AI memory system. And it's free.

career-ops. Open-source AI job search: scan job portals, evaluate listings with a structured A-F rubric into a 1.0-5.0 score, tailor your CV, track applications — runs locally in your AI coding CLI (Claude Code, Codex, OpenCode, Antigravity…)

feynman. TypeScript

ECC. The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

syntheseus. Package for Retrosynthetic Planning

claude-scholar. Semi-automated research assistant for academic research and software development. Supports Claude Code, Codex CLI, Kimi Code CLI, and OpenCode across ideation, coding, experiments, writing, and publication.

AutoResearchClaw. Fully autonomous & self-evolving research from idea to paper. Chat an Idea. Get a Paper. 🦞

autoresearch. AI agents running research on single-GPU nanochat training automatically

AutoFigure-Edit. Python

Paper2Agent. Paper2Agent is a multi-agent AI system that automatically transforms research papers into interactive AI agents with minimal human input.

browser. Lightpanda: the headless browser designed for AI and automation

ChemGraph. Agentic framework for computational chemistry and materials science workflows

jaxsr. A symbolic regression engine for jax

pinchtab. High-performance browser automation bridge and multi-instance orchestrator with advanced stealth injection and real-time dashboard.

frontend-slides. Create beautiful slides on the web using a coding agent's frontend skills

automaton. The first AI that can earn its own existence, replicate, and evolve — without needing a human

AutoFigure. TypeScript

materials_discovery. Jupyter Notebook

langextract. A Python library for extracting structured information from unstructured text using LLMs with precise source grounding and interactive visualization.

scientific-agent-skills. Turn any AI agent into an AI Scientist. The #1 Agent Skills library for science, used by 170,000+ scientists worldwide. 158 ready-to-use skills plus 100+ scientific databases covering biology, chemistry, medicine, and drug discovery. Compatible with Cursor, Claude Code, Codex, Pi, Antigravity, and the open Agent Skills standard.

phonopy. Phonon code

labbench2. labbench2

molfunc. fast functionalisation of molecules

QUASAR. QUASAR is an autonomous system for end-to-end scientific discovery, integrating LLMs with simulation tools to automate workflows across quantum chemistry, materials science, and molecular simulation.

seekpath. A module to obtain and visualize k-vector coefficients and obtain band paths in the Brillouin zone of crystal structures

packmol. Packmol - Initial configurations for molecular dynamics simulations

SynCoGen. Synthesizable 3D Molecule Generation via Joint Reaction and Coordinate Modeling

papersgpt-for-zotero. A powerful Zotero AI and MCP plugin with ChatGPT, Gemini 3.6, Claude Fable 5, Claude Sonnet 5, DeepSeek V4, Grok, OpenRouter, Kimi k3, GLM 5.2, SiliconFlow, GPT-oss, Gemma 4, Qwen 3.7

upet. Universal machine-learning models for advanced atomistic simulations

Generative_AI_For_Science.

course-site-quarto. A Quarto-based course website template

solvOR. Optimization solvers in pure Python: LP, MILP, SAT, constraint programming, graph and metaheuristics. No dependencies. Solvor all your optimization needs.

turm. TUI for the Slurm Workload Manager

PlutoPapers.jl. Interactive and LaTeX-styled papers in Pluto.jl

chemeleon2. A RL framework for Crystal Structure Generation using GRPO

e3nn. A modular framework for neural networks with Euclidean symmetry

aimsgb. Ab-initio Interface Materials Simulation Project for Grain Boundaries (AIMSGB)

atomman-gb. Grain-boundary generator with Atomman

pueue. :stars: Manage your shell commands.

liflow. Flow matching for accelerated simulation of atomic transport

OpenThermo. A comprehensive C++ program for calculating molecular thermochemistry properties from quantum chemistry outputs (Gaussian, Orca, GAMESS, NWCHEM, CP2K, VASP, Q-Chem). OpenThermo implements state-of-the-art methods for computing thermodynamic quantities using statistical mechanics & partition functions.

blip. This repository contains the source code for Bayesian Learned Interatomic Potentials (BLIP)

chemprop. Message Passing Neural Networks for Molecule Property Prediction

regress-lm. Library for sequence-to-sequence numeric prediction, applicable to any tokenizable input, and allows pretraining and fine-tuning over multiple tasks.

SciLink. LLM-powered agents for scientific research automation

MaterialsFramework. A modular Python framework for deploying, benchmarking, and analyzing state-of-the-art machine learning potentials and simulation workflows in materials science.

PhaseForge. PhaseForge is a framework for high-throughput alloy phase diagram prediction using machine learning interatomic potentials (MLIPs) integrated with ATAT and CALPHAD modeling. Includes tools for structure relaxation, molecular dynamics, and TDB generation.

manuscript-template-vscode. TeX

atomify. Real time molecular dynamics in the browser using LAMMPS

rclone-bisync-manager. Python

fipy. FiPy is a Finite Volume PDE solver written in Python

AlgoTune. AlgoTune is a NeurIPS 2025 benchmark made up of 154 math, physics, and computer science problems. The goal is write code that solves each problem, and is faster than existing implementations.

al-folio. A beautiful, simple, clean, and responsive Jekyll theme for academics

synthetic-distillation. Repository for data and figure generation used in "Distillation of atomistic foundation models across architectures and chemical domains"

wrappers. FastAPI wrappers for 3rd-party APIs to format them for integrated use on Ouro

atomistic-software. Tracking citations of atomistic simulation engines

gemini-cli. An open-source AI agent that brings the power of Gemini directly into your terminal.

erwin. Erwin: A Tree-based Hierarchical Transformer for Large-scale Physical Systems [ICML'25]

DistMLIP. DistMLIP: A Distributed Inference Library for Fast, Large Scale Atomistic Simulation

lammps. Public development project of the LAMMPS MD software package

phono3py. A simulation package of phonon-phonon interaction related properties

tblite. Light-weight tight-binding framework

g-xtb. Development versions of the g-xTB method. Final implementation will not happen here but in tblite (https://github.com/tblite/tblite).

augment-atoms. dataset augmentation for atomistic machine learning

make-a-gif. easily make gifs using matplotlib

cif-bond-analyzer. An interactive Python script that computes the minimum atomic bonding distances from sites, generating histograms and pair counts.

Extending-and-Modifying-LAMMPS-Writing-Your-Own-Source-Code. Extending and Modifying LAMMPS Writing Your Own Source Code, published by Packt

janus-core. Tools for machine learnt interatomic potentials

sympy-mcp. A MCP server for symbolic manipulation of mathematical expressions

calphy. A Python library and command line interface for automated free energy calculations

OMatG. Generative framework for crystal structure prediction and de novo generation of inorganic crystals.

Paper2Code. Paper2Code: Automating Code Generation from Scientific Papers in Machine Learning

ReCall. ReSearch: Learning to Reason with Search for LLMs via Reinforcement Learning & ReCall: Learning to Reason with Tool Call for LLMs via Reinforcement Learning

aviary. The Wren sits on its Roost in the Aviary.

mpmorph. MPmorph is a collection of tools to run and analyze ab-initio molecular dynamics (AIMD) calculations run with VASP, and is currently under development. It relies heavily on tools developed by the Materials Project (pymatgen, custodian, fireworks) and atomate.

ATOMICA. Learning Universal Representations of Intermolecular Interactions with ATOMICA

grip. Grand canonical optimization of grain boundary phases.

LESets. A graph neural network for high-entropy alloys.

torch-sim. Torch-native, batchable, atomistic simulations.

trajcast. TrajCast: Force-Free MD Through Autoregressive Equivariant Networks

CellListMap.jl. Flexible implementation of cell lists to map the calculations of particle-pair dependent functions, such as forces, energies, neighbor lists, etc.

k_SRME. Heat-conductivity benchmark test for foundational machine-learning potentials

i-pi. i-PI: a universal force engine

lammps. Compiled binaries and sources of LAMMPS, redistributed by AdvanceSoft Corp.

GPJax. Gaussian processes in JAX and Equinox.

pySED. To implement the phonon SED (spectral energy desity) method in 2010, phonon lifetime can be calculated.

GoogleGenAI.jl. Unofficial Google Gen AI Julia SDK

metatrain. Train, fine-tune, and manipulate machine learning models for atomistic systems

metatrain. Training and evaluating machine learning models for atomistic systems.

awesome-cursorrules. 📄 Configuration files that enhance Cursor AI editor experience with custom rules and behaviors

jrystal. A JAX-based Differentiable Density Functional Theory Framework for Materials

matten. MatTen: Equivariant Graph Neural Nets for Tensorial Properties of Materials

moyo. Library for Crystal Symmetry in Rust

mace-foundations. MACE foundation models (MP, OMAT, mh-1)

MLFF-distill. distill repo with submoduled fairchem

PLAMS. Python Library for Automating Molecular Simulations

pml-book. "Probabilistic Machine Learning" - a book series by Kevin Murphy

torchani. TorchANI 2.0 is an open-source library that supports training, development, and research of ANI-style neural network interatomic potentials. It was originally developed and is currently maintained by the Roitberg group.

schnetpack. SchNetPack - Deep Neural Networks for Atomistic Systems

PLACER. PLACER is graph neural network for local prediction of protein-ligand conformational ensembles.

RDsynthesisML. This repository contains python scripts and notebooks for replicating the results in "Towards Autonomous Materials Synthesis via Reaction-Diffusion Coupling".

matid. MatID is a Python package for identifying and analyzing atomistic systems based on their structure.

dspy. DSPy: The framework for programming—not prompting—language models

lerobot. 🤗 LeRobot: Making AI for Robotics more accessible with end-to-end learning

MACE-OFF-mu-models. A dipole model trained on the SPICE dataset.

lematerial-hasher. Python

deep-research. An AI-powered research assistant that performs iterative, deep research on any topic by combining search engines, web scraping, and large language models. The goal of this repo is to provide the simplest implementation of a deep research agent - e.g. an agent that can refine its research direction overtime and deep dive into a topic.

autofd. Automatic Functional Differentiation in JAX

NeuralForceField. Neural Network Force Field based on PyTorch

pymatgen-analysis-diffusion. This add-on to pymatgen provides tools for analyzing diffusion in materials.

devin.cursorrules. Magic to turn Cursor/Windsurf as 90% of Devin

oumi. Easily fine-tune, evaluate and deploy Qwen, Gemma, or any open weight LLM!

qwen-web. qwen running locally in your browser

digital-experiments. 🧪👨‍💻 keep track of your python experiments

simpleRL-reason. Simple RL training for reasoning

BayesianOptimization. A Python implementation of global optimization with gaussian processes.

openai-realtime-agents. This is a simple demonstration of more advanced, agentic patterns built on top of the Realtime API.

ml_uncertainty. Get prediction intervals, confidence intervals, and parameter uncertainties for various machine learning models

mattergen. Official implementation of MatterGen -- a generative model for inorganic materials design across the periodic table that can be fine-tuned to steer the generation towards a wide range of property constraints.

cace. Python

aimnetcentral. AIMNet2: Fast and accurate machine-learned interatomic potential for molecular dynamics simulations

molbokeh. A new python package to visualize molecules in dots hover

khoj. Your AI second brain. Self-hostable. Get answers from the web or your docs. Build custom agents, schedule automations, do deep research. Turn any online or local LLM into your personal, autonomous AI (gpt, claude, gemini, llama, qwen, mistral). Get started - free.

browser-use. 🌐 Make websites accessible for AI agents. Automate tasks online with ease.

quacc. quacc is a flexible platform for computational materials science and quantum chemistry that is built for the big data era.

Composer. Agentic AI sandbox

marimo-gh-pages-template. A template repo for deploying marimo WebAssembly notebooks to GitHub Pages

litdb. A literature database

MatFold. Package for systematic insights into materials discovery models’ performance through standardized chemical cross-validation protocols

snakemake. This is the development home of the workflow management system Snakemake. For general information, see

mace-osaka24. MACE_Osaka24 models

Scientific-Inkscape. Scientific Inkscape: Inkscape extensions for figure resizing and editing

xtb. Semiempirical Extended Tight-Binding Program Package

markitdown. Python tool for converting files and office documents to Markdown.

exo. Run frontier AI locally.

matpes. A foundational potential energy dataset for materials

mic-hackathon. HTML

python-genai. Google Gen AI Python SDK provides an interface for developers to integrate Google's generative models into their Python applications.

ComFiT. A python library for simulating field theories with topological defects

MINov. A parameter-free approach for estimating materials novelty along chemical and structural axes.

CrysText. CrysText: A Generative AI Approach for Text-Conditioned Crystal Structure Generation using LLM

materials. Foundation Model for Materials - FM4M

graph-pes. train and use graph-based ML models of potential energy surfaces

mlipx. Machine-Learned Interatomic Potential eXploration (mlipx) is designed at BASF for evaluating machine-learned interatomic potentials (MLIPs). It offers a growing set of evaluation methods alongside powerful visualization and comparison tools.

MegaParse. File Parser optimised for LLM Ingestion with no loss 🧠 Parse PDFs, Docx, PPTx in a format that is ideal for LLMs.

mattersim. MatterSim: A deep learning atomistic model across elements, temperatures and pressures.

synthesis_bots_public. Main code for planning, executing, and analysing experiments performed in the synthesis bots autonomous chemistry workflows.

chemeleon. A text-guided diffusion model for crystal structure generation

awesome-robot-descriptions. A curated list of awesome robot descriptions (URDF, MJCF)

brutus. Julia

open-notebooklm. Convert any PDF into a podcast episode!

autoWTE. Benchmarking foundation Machine Learning Potentials with Lattice Thermal Conductivity from Anharmonic Phonons

cetz-plot. Create Plots and Charts with CeTZ

aviary. A language agent gym with challenging scientific tasks

claude-cookbooks. A collection of notebooks/recipes showcasing some fun and effective ways of using Claude.

autoplex. Code for automated fitting of machine learned interatomic potentials.

swarm. Educational framework exploring ergonomic, lightweight multi-agent orchestration. Managed by OpenAI Solution team.

rip2. A safe and ergonomic alternative to rm

BitNet. Official inference framework for 1-bit LLMs

fairchem. FAIR Chemistry's library of machine learning methods for chemistry

outlines. Structured Outputs

SEPIA. Simulation-Enabled Prediction, Inference, and Analysis: physics-informed statistical learning.

so3lr. SO3krates and Universal Pairwise Force Field for Molecular Simulation

GNNOpt. Universal Ensemble-Embedding Graph Neural Network for Direct Prediction of Optical Spectra from Crystal Structures

ForceConstants.jl. Code to calculate force constants of empirical potentials in Julia.

ChemOS2.0. Python

slydst. Create simple static slides with Typst

Darwin. An open-source project dedicated to build foundational large language model for natural science, mainly in physics, chemistry and material science.

quarto-cv. A quarto extensions for a CV.

LLMs-from-scratch. Implement a ChatGPT-like LLM in PyTorch from scratch, step by step

NN-SVG. Publication-ready NN-architecture schematics.

PDF2Audio. Jupyter Notebook

Crystalyze. Python

MolPipeline. A Python package for processing molecules with RDKit in scikit-learn

ReinforcementLearningAnIntroduction.jl. Julia code for the book Reinforcement Learning An Introduction

cinelog. Comprehensive logging of all terminal input and output for each session based on Asciinema and wild zsh + Python scripting.