Jan Hermann

Computational chemist and physicist by training️, I later became hooked on machine learning and now focus on integrating deep learning into quantum chemistry. I was born and raised in Český Krumlov, moved to Prague for university, and later to Berlin for a Phd, where I live with my wife and our two kids. I like cycling, coffee, and chess. I work at Microsoft in AI for Science.

research: Google Scholar · ORCID
code: GitHub
social: Bluesky · LinkedIn
cycling: Strava
books: Goodreads
chess: Lichess

Works

I build quantum-chemistry tools that others build on, tracing an arc from hand-crafted physics modeling of long-range correlation, to physics-driven machine learning of wavefunctions, to data-driven machine learning of the exchange–correlation functional.

Density functional theory

Skala · contributor · 240
Skala exchange-correlation functional (Python)

Exchange–correlation, learned from data. Skala learns the exchange–correlation functional of Kohn–Sham DFT, reaching hybrid-functional accuracy at semi-local cost.¹ This thread started by looking at the algorithmic side of things,² but it later became evident that the key ingredient is data.^3,4

1	Accurate and scalable exchange-correlation with deep learning · G. Luise et al. · Preprint at arXiv:2506.14665 (2025)	24
2	Variational principle to regularize machine-learned density functionals: The non-interacting kinetic-energy functional · P. del Mazo-Sevillano & JH · J. Chem. Phys. 159, 194107 (2023)	17
3	Accurate Chemistry Collection: Coupled cluster atomization energies for broad chemical space · S. Ehlert et al. · Sci. Data (2026)	5
4	Chemical Space Exploration with Artificial “Mindless” Molecules · T. Gasevic, M. Müller, J. Schöps, S. Lanius, JH, S. Grimme & A. Hansen · J. Chem. Inf. Model. 65, 9576–9587 (2025)	5

Talks (1)

2025	“Skala: Accurate and scalable exchange-correlation with deep learning” · Symposium on Theoretical Chemistry (Berlin, Germany)

Quantum Monte Carlo

DeepQMC · creator · 416
Deep learning quantum Monte Carlo for electrons in real space (Python)

Wavefunctions, learned from physics. Deep QMC began with PauliNet,⁵ the first deep-learning ansatz to reach chemical accuracy through variational Monte Carlo — the wavefunction optimized against the Schrödinger equation itself, with no reference data. It can be converged to the fixed-node limit,⁶ reach excited states,⁷ and eventually grew into a general open-source suite⁸ within a fast-developing field.⁹ Those same wavefunctions also yield accurate real-space electron densities,¹⁰ and in the latest iteration are encoded in a single foundation model.¹¹

5	Deep-neural-network solution of the electronic Schrödinger equation · JH, Z. Schätzle & F. Noé · Nat. Chem. 12, 891–897 (2020)	857
6	Convergence to the fixed-node limit in deep variational Monte Carlo · Z. Schätzle, JH & F. Noé · J. Chem. Phys. 154, 124108 (2021)	33
7	Electronic excited states in deep variational Monte Carlo · M. T. Entwistle, Z. Schätzle, P. A. Erdman, JH & F. Noé · Nat. Commun. 14, 274 (2023)	95
8	DeepQMC: An open-source software suite for variational optimization of deep-learning molecular wave functions · Z. Schätzle, P. B. Szabó, M. Mezera, JH & F. Noé · J. Chem. Phys. 159, 094108 (2023)	41
9	Ab initio quantum chemistry with neural-network wavefunctions · JH, J. Spencer, K. Choo, A. Mezzacapo, W. M. C. Foulkes, D. Pfau, G. Carleo & F. Noé · Nat. Rev. Chem. 7, 692–709 (2023)	204
10	Highly accurate real-space electron densities with neural networks · L. Cheng, P. B. Szabó, Z. Schätzle, D. P. Kooi, J. Köhler, K. J. H. Giesbertz, F. Noé, JH, P. Gori-Giorgi & A. Foster · J. Chem. Phys. 162, 034120 (2025)	19
11	An ab initio foundation model of wavefunctions that accurately describes chemical bond breaking · A. Foster, Z. Schätzle, P. B. Szabó, L. Cheng, J. Köhler, G. Cassella, N. Gao, J. Li, F. Noé & JH · Preprint at arXiv:2506.19960 (2025)	22

Talks (12)

2024	“Neural-network wave functions for quantum chemistry” · European Seminar on Computational Methods in Quantum Chemistry (Copenhagen, Denmark)
2023	“Solving the electronic Schrödinger equation with deep learning” · SIAM Conference on Computational Science and Engineering (Amsterdam, Netherlands)
2022	“Neural-network wave functions for quantum chemistry” · MLQC4DYN (Institut Pascal, Paris, France)
	“Neural-network wave functions for quantum chemistry” · Monte Carlo and Machine Learning Approaches in Quantum Mechanics (IPAM, Los Angeles, USA)
2021	“Deep-learning solution to the electronic many-body problem” · Non-Covalent Interactions in Large Molecules and Extended Materials (EPFL, Lausanne, Switzerland)
	“Solving the electronic Schrödinger equation with deep learning” · ACS Fall Meeting [virtual]
	“Approaching exact solutions of the electronic Schrödinger equation with deep quantum Monte Carlo” · APS March Meeting [virtual]
	“Solving the electronic Schrödinger equation with deep learning” · Stochastic Methods in Electronic Structure Theory [virtual]
2020	“Deep neural network solution of the electronic Schrödinger equation” · APS March Meeting (Denver, USA) [cancelled]
	“Convergence to the fixed-node limit in deep variational Monte Carlo” · NeurIPS workshop Machine Learning and the Physical Sciences [virtual]
	“Solving the electronic Schrödinger equation with deep learning” · Scientific Machine Learning Mini-Course (Carnegie Mellon University) [virtual]
2019	“Deep neural network solution of the electronic Schrödinger equation” · NeurIPS workshop Machine Learning and the Physical Sciences (Vancouver, Canada)

Van der Waals dispersion

libMBD · creator · 61
Many-body dispersion library (Fortran)

Long-range electron correlation, by hand. The itch started with a vdW-DF/CCSD(T) correction scheme^12,13 for zeolites^14,15 and grew into a unified density-functional model of van der Waals interactions,^16–18 with the exchange–correlation balance worked out along the way.^19,20 Packaged as libMBD, a scalable many-body dispersion library²¹ now embedded in several electronic-structure codes, it underpins applications from π–π stacked molecules²² through molecular crystals and layered materials^23–26 to Casimir and fluctuational-electrodynamics phenomena.^27–31

12	A novel correction scheme for DFT: A combined vdW-DF/CCSD(T) approach · JH & O. Bludský · J. Chem. Phys. 139, 034115 (2013) * *This article may be downloaded for personal use only. Any other use requires prior permission of the author and AIP Publishing.	20
13	Nonlocal correlation in density functional theory · JH · Charles University (2013)
14	Theoretical investigation of layered zeolite frameworks: Surface properties of 2D zeolites · JH, M. Trachta, P. Nachtigall & O. Bludský · Catal. Today 227, 2–8 (2014)	27
15	Theoretical investigation of the Friedländer reaction catalysed by CuBTC: Concerted effect of the adjacent Cu²⁺ sites · M. Položij, E. Pérez-Mayoral, J. Čejka, JH & P. Nachtigall · Catal. Today 204, 101–107 (2013)	41
16	Density functional model for van der Waals interactions: Unifying many-body atomic approaches with nonlocal functionals · JH & A. Tkatchenko · Phys. Rev. Lett. 124, 146401 (2020)	199
17	Towards unified density-functional model of van der Waals interactions · JH · Humboldt University (2018)	6
18	First-principles models for van der Waals interactions in molecules and materials: Concepts, theory, and applications · JH, R. A. DiStasio, Jr. & A. Tkatchenko · Chem. Rev. 117, 4714–4758 (2017) * *This document is the unedited Author’s version of a Submitted Work that was subsequently accepted for publication in Chemical Reviews, copyright © American Chemical Society after peer review. To access the final edited and published work follow this link.	776
19	Electronic exchange and correlation in van der Waals systems: Balancing semilocal and nonlocal energy contributions · JH & A. Tkatchenko · J. Chem. Theory Comput. 14, 1361–1369 (2018) * *This document is the unedited Author’s version of a Submitted Work that was subsequently accepted for publication in Journal of Chemical Theory and Computation, copyright © American Chemical Society after peer review. To access the final edited and published work follow this link.	42
20	Van der Waals interactions in material modelling · JH & A. Tkatchenko · In Handbook of materials modeling (eds W. Andreoni & S. Yip) 1–33 (Springer, 2018)	4
21	libMBD: A general-purpose package for scalable quantum many-body dispersion calculations · JH, M. Stöhr, S. Góger, S. Chaudhuri, B. Aradi, R. J. Maurer & A. Tkatchenko · J. Chem. Phys. 159, 174802 (2023)	27
22	Nanoscale π–π stacked molecules are bound by collective charge fluctuations · JH, D. Alfè & A. Tkatchenko · Nat. Commun. 8, 14052 (2017)	122
23	Communication: Many-body stabilization of non-covalent interactions: Structure, stability, and mechanics of Ag₃Co(CN)₆ framework · X. Liu, JH & A. Tkatchenko · J. Chem. Phys. 145, 241101 (2016) * *This article may be downloaded for personal use only. Any other use requires prior permission of the author and AIP Publishing.	19
24	Tuning intermolecular interactions with nanostructured environments · M. Chattopadhyaya, JH, I. Poltavsky & A. Tkatchenko · Chem. Mater. 29, 2452–2458 (2017) * *This document is the unedited Author’s version of a Submitted Work that was subsequently accepted for publication in Chemistry of Materials, copyright © American Chemical Society after peer review. To access the final edited and published work follow this link.	12
25	Nonlocal electronic correlations in the cohesive properties of high-pressure hydrogen solids · T. Cui, J. Li, W. Gao, JH, A. Tkatchenko & Q. Jiang · J. Phys. Chem. Lett. 11, 1521–1527 (2020) * *This document is the unedited Author’s version of a Submitted Work that was subsequently accepted for publication in The Journal of Physical Chemistry Letters, copyright © American Chemical Society after peer review. To access the final edited and published work follow this link.	10
26	Anisotropic interlayer force field for transition metal dichalcogenides: The case of molybdenum disulfide · W. Ouyang, R. Sofer, X. Gao, JH, A. Tkatchenko, L. Kronik, M. Urbakh & O. Hod · J. Chem. Theory Comput. 17, 7237–7245 (2021)	56
27	Unifying microscopic and continuum treatments of van der Waals and Casimir interactions · P. S. Venkataram, JH, A. Tkatchenko & A. W. Rodriguez · Phys. Rev. Lett. 118, 266802 (2017) * *Copyright 2017 by the American Physical Society	35
28	Phonon-polariton mediated thermal radiation and heat transfer among molecules and macroscopic bodies: Nonlocal electromagnetic response at mesoscopic scales · P. S. Venkataram, JH, A. Tkatchenko & A. W. Rodriguez · Phys. Rev. Lett. 121, 045901 (2018) * *Copyright 2018 by the American Physical Society	19
29	Impact of nuclear vibrations on van der Waals and Casimir interactions at zero and finite temperature · P. S. Venkataram, JH, T. J. Vongkovit, A. Tkatchenko & A. W. Rodriguez · Sci. Adv. 5, eaaw0456 (2019)	9
30	Fluctuational electrodynamics in atomic and macroscopic systems: van der Waals interactions and radiative heat transfer · P. S. Venkataram, JH, A. Tkatchenko & A. W. Rodriguez · Phys. Rev. B 102, 085403 (2020) * *Copyright 2020 by the American Physical Society	4
31	Coulomb interactions between dipolar quantum fluctuations in van der Waals bound molecules and materials · M. Stöhr, M. Sadhukhan, Y. S. Al-Hamdani, JH & A. Tkatchenko · Nat. Commun. 12, 137 (2021)	47

Talks (24)

2022	“Libmbd: A general-purpose package for scalable many-body dispersion calculations” · Electronic Structure Software Development (Lausane, Switzerland) [virtual]
2020	“Density-functional model for van der Waals interactions: Unifying atomic approaches with nonlocal functionals” · Electronic Structure Theory with Numeric Atom-Centered Basis Functions [virtual]
2019	“Unifying density-functional and interatomic approaches to van der Waals interactions” · Frontiers in Density Functional Theory and Beyond (Kavli ITS, Beijing, China)
2018	“Modeling van der Waals interactions in molecules and materials” · Molecular Simulations Meets Machine Learning and Artificial Intelligence (Lorentz Center, Leiden, Netherlands)
	“Modeling van der Waals interactions in materials with many-body dispersion” · Electronic Structure Theory with Numeric Atom-Centered Basis Functions (TU Munich, Germany)
	“Modeling van der Waals interactions” · Python for Quantum Chemistry and Materials Simulation Software (Caltech, Pasadena, USA)
	“Unified many-body approach to van der Waals interactions based on semilocal polarizability functional” · APS March Meeting (Los Angeles, USA)
2017	“What is the range of electron correlation in density functionals?” · APS March Meeting (New Orleans, USA)
	“Balancing semilocal and nonlocal energy contributions in van der Waals systems” · Intermolecular Interactions (Arenas de Cabrales, Spain)
2016	“First-principles approaches to van der Waals interactions” · Many-Body Interactions (Telluride, USA)
	“Nanoscale π–π stacked molecules bound by collective charge fluctuations” · Aspuru-Guzik group seminar (Harvard University, Cambridge, USA)
2015	“Many-body dispersion meets non-local density functionals” · Modeling Many-Body Interactions (Lake La Garda, Italy)
	“Many-body dispersion meets non-local density functionals” · DPG March Meeting (Berlin, Germany)
	“Many-body dispersion meets non-local density functionals” · APS March Meeting (San Antonio, USA)
	“Non-local density functionals meet many-body dispersion” · Psi-k Conference (San Sebastian, Spain)
	“Many-body dispersion meets non-local density functionals” · Congress of Theoretical Chemists (Torino, Italy)
	“Non-local density functionals meet many-body dispersion” · Frontiers of First-Principles Simulations: Materials Design and Discovery (Berlin, Germany)
2014	“Non-local density functionals meet many-body dispersion” · DPG March Meeting (Dresden, Germany)
	“Non-local density functionals meet many-body dispersion” · Addressing Challenges for First-Principles Based Modeling of Molecular Materials (Lausanne, Switzerland)
2013	“Adsorption in zeolites investigated by dispersion-corrected DFT” · Layered Materials (Liblice, Czechia)
	“Modeling of surface properties of lamellar zeolites” · Molecular Sieves (Heyrovsky Institute, Prague, Czechia)
	“Modeling of surface properties of lamellar zeolites” · Molecular Sieves and Catalysis (Segovia, Spain)
2012	“Silver clusters in zeolites: Structure, stability and photoactivity” · British Zeolite Association Meeting (Chester, UK)
	“Silver clusters in faujasite: A theoretical investigation” · Molecular Sieves (Prague, Czechia)

Ecosystem

Pyberny · creator · 130
Molecular structure optimizer (Python)
FHI-aims · core contributor
All-electron electronic-structure calculations (Fortran)
PySCF · contributor
DFTB+ · contributor
QCEngine · contributor

No tool stands alone. libMBD is built into FHI-aims, DFTB+, and other electronic-structure codes; Pyberny into PySCF and QCEngine — integrations reflected in their program papers.^32–35 The same connective work runs through survey writing across the field, from a roadmap on machine learning in electronic structure³⁶ to an introduction to material modeling,³⁷ alongside the invited seminars below.

32	Roadmap on Advancements of the FHI-aims Software Package · J. W. Abbott et al. · Preprint at arXiv:2505.00125 (2025)	23
33	DFTB+, a software package for efficient approximate density functional theory based atomistic simulations · B. Hourahine et al. · J. Chem. Phys. 152, 124101 (2020) * *This article may be downloaded for personal use only. Any other use requires prior permission of the author and AIP Publishing.	1343
34	Recent developments in the PʏSCF program package · Q. Sun et al. · J. Chem. Phys. 153, 024109 (2020) * *This article may be downloaded for personal use only. Any other use requires prior permission of the author and AIP Publishing.	1536
35	Quantum Chemistry Common Driver and Databases (QCDB) and Quantum Chemistry Engine (QCEɴɢɪɴᴇ): Automation and interoperability among computational chemistry programs · D. G. A. Smith et al. · J. Chem. Phys. 155, 204801 (2021)	59
36	Roadmap on Machine learning in electronic structure · H. J. Kulik et al. · Electron. Struct. 4, 023004 (2022)	228
37	Introduction to material modeling · JH · In Machine learning meets quantum physics (eds K. T. Schütt et al.) 7–24 (Springer, 2020)

Talks (12)

2022	UCT & IOCB Theoretical Chemistry Seminar (VŠCHT, Prague, Czechia)
	Lennard-Jones Centre Discussion Group (University of Cambridge) [virtual]
2021	Molecular and Ultrafast Science Seminar (Center for Free-Electron Laser Science) [virtual]
	Machine Learning seminar (Chalmers University of Technology) [virtual]
	Grüneis group seminar (TU Wien) [virtual]
	(Nano)Materials Modeling Seminar (Charles University) [virtual]
	Cosmology Seminar (University of Szczecin) [virtual]
2020	Machine Learning in Physics, Chemistry and Materials (University of Cambridge) [virtual]
	Jordan group seminar (University of Pittsburgh) [virtual]
2018	“Mona: Calculation framework for reproducible science” · Theory Department seminar (Fritz Haber Institute, Berlin, Germany)
2016	“Python interface to FHI-aims” · Electronic Structure Theory with Numeric Atom-Centered Basis Functions (Munich, Germany)
2015	DiStasio group seminar (Cornell University, Ithaca, USA)

Employment

	Microsoft, Berlin
Nov 2022–	Principal research manager · AI for Science
	Free University of Berlin
Nov 2020–Oct 2022	Junior research group leader · Department of Mathematics
Jan 2019–Oct 2020	Postdoctoral researcher · AI4Science group
	University of Luxembourg
Jan–Dec 2018	Postdoctoral researcher · Theoretical Chemical Physics group
	Fritz Haber Institute, Berlin
Oct 2013–Dec 2017	Graduate research assistant · Theory department
	Institute of Organic Chemistry and Biochemistry, Prague
Mar 2010–Sep 2013	Undergraduate research assistant · Non-Covalent Interactions group

Education

	Humboldt University of Berlin
Dec 2017	Ph.D. in Physics · summa cum laude
	Charles University, Prague
Sep 2013	M.S. in Molecular Modeling
Sep 2011	B.S. in Physics
Jun 2011	B.S. in Chemistry

Secondary appointments

Jul 2021–Oct 2022	Junior Fellow · BIFOLD, Berlin
Jan 2019–Oct 2020	Postdoctoral research fellow · Machine Learning group, TU Berlin
Sep–Dec 2016	Visiting graduate researcher · IPAM, UCLA (long program “Understanding Many-Particle Systems with Machine Learning”)

Awards

Feb 2021	Marie Skłodowska-Curie Individual Fellowship [relinquished]
Jan 2014	Heyrovsky Prize for the best science graduate · Charles University
Jul 2008	Gold Medal · 39th International Physics Olympiad

Professional activities

Peer-reviewed 43 manuscripts for Phys. Rev. X, Nat. Commun., Nat. Mach. Intell., Phys. Rev. Lett., J. Chem. Phys., and other journals
Reviewed 1 grant proposal for U. S. Department of Energy