← home

Janosh Riebesell

I joined the Materials Project in early 2023 where I developed ML foundation models (CHGNet, MACE-MP) and build high-throughput workflows for generating large DFT datasets to train still bigger models. I am a co-maintainer of pymatgen. I'm a big fan of high-quality open source software with a focus on enabling new capabilities for scaling computational materials science. GitHub is where most of my work happens.

  Selected Publications Sort by

  1. Developments and applications of the OPTIMADE API for materials discovery, design, and data exchange

    M. Evans, ..., J. Riebesell, ..., R. Armiento — 10.48550/arXiv.2402.00572 — 2024-2

  2. LLaMP: Large Language Model Made Powerful for High-fidelity Materials Knowledge Retrieval and Distillation

    Y. Chiang, C. Chou, J. Riebesell — arxiv.org/abs/2401.17244 — 2024-1

  3. Pushing the Pareto front of band gap and permittivity: ML-guided search for dielectric materials

    J. Riebesell, T. Surta, ..., A. Lee — arxiv.org/abs/2401.05848v1 — 2024-1

  4. Jobflow: Computational Workflows Made Simple

    A. Rosen, ..., J. Riebesell, ..., A. Ganose — 10.21105/joss.05995 — 2024-1

  5. A foundation model for atomistic materials chemistry

    J. Riebesell, I. Batatia, ..., G. Csányi — arxiv.org/abs/2401.00096v1 — 2023-12

  6. CHGNet as a pretrained universal neural network potential for charge-informed atomistic modelling

    B. Deng, ..., J. Riebesell, ..., G. Ceder — 10.1038/s42256-023-00716-3 — 2023-9

  7. Matbench Discovery - An evaluation framework for machine learning crystal stability prediction

    J. Riebesell, R. Goodall, ..., A. Lee — 10.48550/arXiv.2308.14920 — 2023-8

  8. Crystal Toolkit: A Web App Framework to Improve Usability and Accessibility of Materials Science Research Algorithms

    M. Horton, ..., J. Riebesell, ..., K. Persson — 10.48550/arXiv.2302.06147 — 2023-2

  Open Source Sort by

  • pymatgen Logo pymatgen 1336 ⭐ 979 commits Python, Cython, Jupyter Notebook

    One of the largest and most popular open source materials analysis codes that defines classes for structures, molecules, slabs, etc. and interfaces seamlessly with various other materials codes. It also powers the Materials Project.

  • atomate2 Logo atomate2 119 ⭐ 357 commits Python

    atomate2 is a library of computational materials science workflows used by the Materials Project and beyond. It supports multiple DFT codes and downstream analysis tools. Recently, we added machine learning potential-powered structure relaxation workflows.

  • Matbench Discovery Logo Matbench Discovery 66 ⭐ 318 commits Python, Svelte, CSS

    Benchmark for machine learning energy models simulating a real-world materials discovery campaign.

  • MultiSelect Logo MultiSelect 263 ⭐ 271 commits TypeScript, Svelte, CSS

    Keyboard-friendly, accessible and customizable multi-select web component.

  • pymatviz Logo pymatviz 111 ⭐ 242 commits Python, Svelte, CSS

    A toolkit for visualizations in materials informatics to complement pymatgen.

  • Aviary Logo Aviary 39 ⭐ 234 commits Python

    Compositional, structural and coarse-grained structural ML energy model implementations (Roost, Wren, CGCNN, Wrenformer) with a consistent API.

  • TikZ Logo TikZ 174 ⭐ 187 commits TeX, Svelte, Python

    Collection TikZ figures for concepts in physics/chemistry/ML.

  • CHGNet Logo CHGNet 184 ⭐ 175 commits Python, C, Cython

    Pretrained universal neural network potential for charge-informed atomistic modeling published on the Sep 2023 cover of NMI.

  • Elementari Logo Elementari 118 ⭐ 174 commits TypeScript, Svelte, CSS

    A library of Svelte components for building interactive web apps with performant chemistry visualizations like periodic tables, Bohr atoms, nuclei, heatmaps, scatter plots.

  • jobflow Logo jobflow 83 ⭐ 115 commits Python, TeX

    jobflow is a library for writing computational workflows. It provides the plumbing underlying atomate2 and was adopted by several other workflow libraries.

  • Tensorboard Reducer Logo Tensorboard Reducer 65 ⭐ 79 commits Python, TeX

    Reduce multiple PyTorch TensorBoard runs to new events/CSV/JSON. Good for model ensembles.

  • Normalizing Flows Logo Normalizing Flows 1283 ⭐ 72 commits Python

    Curated list of resources for learning and using normalizing flows, a powerful tool in ML for modeling probability distributions.

  • MatCalc Logo MatCalc 41 ⭐ 72 commits Python

    A Python library for calculating materials properties from ML force field potential energy surfaces.

  • MACE Logo MACE 345 ⭐ 21 commits Python, Shell

    Fast and accurate machine learning interatomic potentials with higher order equivariant message passing.

  • Dielectrics Logo Dielectrics 7 ⭐ 14 commits Python, ReScript, Roff

    Pushing the Pareto front of band gap and permittivity with ML-guided dielectrics discovery incl. experimental synthesis.

  Education

  • PhD Student University of Cambridge

    Thesis title: Can machine learning accelerate high-throughput searches for novel functional materials?

  • MPhil in Scientific Computing University of Cambridge

    Thesis title: Probabilistic Data-Driven Discovery of Thermoelectric Materials

  • MSc in Physics ITP Heidelberg

    Thesis title: Functional Renormalization Group Analytically Continued to Finite Temperatures

  • BSc in Physics Hamburg University

    Thesis title: van der Waals Corrections for Density Functional Theory - DFT+D2 applied to Graphene-hBN-Heterostructures

  Awards

  Volunteer Work

  • Studenten bilden SchülerStudenten bilden Schüler Board member and head of IT

    Non-profit student initiative that connects University student volunteers with children from underprivileged families for free tutoring.

  • Afara FoundationAfara Foundation Board member and head of IT

    A non-profit student association that runs orphanages in Namibia and funds children's education and medical treatment in several African countries.