MPMichael F. Peintinger

Michael F. Peintinger

*This is an algorithmically generated author record based on publicly available information.

Michael F. Peintinger has a total of 15 co-authors. Here are the top co-authors by number of shared publications:

  • Thomas Bredow

    Mulliken Center for Theoretical Chemistry

    4 shared publications
  • Bartolomeo Civalleri

    Department of Chemistry University of Torino Torino Italy

    1 shared publication
  • Daniel Vilela Oliveira

    Mulliken Center for Theoretical Chemistry, Institut für Physikalische und Theoretische Chemie, University of Bonn, Beringstr. 4, Bonn D-53115, Germany

    1 shared publication
  • Florian Janetzko

    Theoretische Chemie, Universität Hannover, Am Kleinen Felde 30, 30167 Hannover, Germany

    1 shared publication
  • Jan Gerit Brandenburg

    Mulliken Center for Theoretical Chemistry, Institut für Physikalische und Theoretische Chemie der Universität Bonn , Beringstraße 4, 53115 Bonn, Germany

    1 shared publication
  • Johannes Beck

    Institute for Inorganic Chemistry, University of Bonn, Bonn, Germany

    1 shared publication

Latest Publications

  • Open Access

    vibyverse: an integrated computational chemistry ecosystem developed through agentic engineering

    by Michael F. Peintinger

    We describe the vibyverse, developed by agentic engineering, comprising vibe-sketcher for chemical structures and reaction schemes, vibe-qc for electronic-structure calculations, vibe-scheduler for execution and retrieval, vibe-viewer for inspection and presentation, and the Quantum Visualization Format (QVF) for structured exchange. The workflow starts with reaction design in vibe-sketcher, submission through vibe-scheduler, calculation with vibe-qc, and inspection and analysis in vibe-viewer, with scientific data transferred through QVF. We examine the responsibilities of each component and the information that must survive their connections, including charge, multiplicity, geometry, method settings, runtime identity and result provenance. Source inspection identifies implemented interfaces and their compatibility requirements. A Diels–Alder preparation example connects an editable scheme and preoptimized structures to validated QVF and viewer exports. The executed numerical workflow proceeds through MACE geometry optimization, PBE/def2-SVP refinement and DLPNO-CCSD(T) endpoint energies. A proposed water-on-boronnitride example would extend the workflow to periodic structures. The molecular optimization and Hessian use existing PBE routes without an added dispersion correction. Both Hessian stencils identify minima for all three molecules. With an explicit restrictedreference contract, the correlated electronic energy of cyclohexene relative to butadiene and ethene is −190.547 kJ mol−1. Changing from TightPNO to NormalPNO shifts it by +0.187 kJ mol−1. The viewer reproduces both electronic reaction energies from the archived endpoints, in agreement with an independent stoichiometric sum. All data supporting this study are openly available from the Zenodo repository under accession number 23976098 at https://zenodo.org/records/23976098.

  • Open Access

    vibe-qc: expert-directed agentic engineering for scientific software development

    by Michael F. Peintinger

    Generative artificial intelligence (AI) can implement scientific software, but useful development also requires executable comparisons, inspectable results and a record of scientific decisions. This case study examines the tooling that enables semi-autonomous electronic-structure software development under expert direction. VIBE-QC supplies molecular and periodic calculations; vibe-scheduler queues and manages calculations across clusters and a fleet of workstations; vibe-viewer and the Quantum Visualization Format (QVF) connect numerical outputs to portable inspection and reproducible figures. Retained development records illustrate how these tools preserve failed attempts, distinguish changes in physical conventions from implementation errors, and separate measured differences from expert-selected acceptance criteria. Publication-specific calculations were run with VIBE-QC 0.18.x and use formaldehyde to connect controlled numerical comparisons with visual inspection; diamond and lithium hydride provide periodic Perdew–Burke–Ernzerhof (PBE) examples with explicit convergence checks. The expert sets the physical problem, selects diagnostic controls and interprets the evidence; AI agents implement and execute those tasks. The resulting account identifies practical requirements for sustained AI-assisted scientific development. It establishes a traceable process within one project, without claiming autonomous scientific judgment or a measured productivity advantage.

  • Open Access

    vibe‐qc : An Open‐Source, General‐Purpose Electronic‐Structure Code for Molecules and Solids Developed With Generative AI

    by Michael F. Peintinger

    We present vibe‐qc , a general‐purpose open‐source electronic‐structure code for molecules and solids, developed almost entirely with agentic AI under expert human direction. The code is released under the Mozilla Public License 2.0 (MPLv2). The implementation comprises a C++17 numerical core and Python calculation drivers. Molecular methods include Hartree–Fock, Kohn–Sham DFT, Møller–Plesset theory and double hybrids, coupled cluster, active‐space methods, and excited states. Dispersion corrections, continuum solvation, and effective core potentials are also available. Periodic calculations use Gaussian‐basis HF and DFT with Brillouin‐zone sampling. Molecular and periodic method coverage differs; the feature matrices specify analytic and numerical derivatives and distinguish validated and experimental implementations. The validation set is a key contribution of this work. It combines independent program comparisons, published benchmarks, and internal consistency tests. Molecular mean‐field energies agree at the nanohartree scale, and conventional CCSD(T) energies agree within for the tested systems. Local‐correlation and periodic limitations are reported with the comparisons. Numerical results, calculation metadata, and analysis scripts are supplied as machine‐readable . The independently maintained vibe‐view viewer, vibe‐queue scheduler, and qvf toolkit support visualization and job management; trexio provides wavefunction interchange.

  • Open Access

    vibe-queue: provenance-aware job execution for computational chemistry across workstations and clusters

    by Michael F. Peintinger

    Computational chemistry campaigns span workstations and institutional clusters, while calculation software changes during development. We present vibe-queue 1.0.0, an open-source execution layer connecting inputs, resource requests, runtime identity, execution state, and recovered outputs. Background services run local and SSH-accessible jobs; external drivers submit cluster jobs through Torque/PBS or Slurm. Fixed runtime environments preserve selected software across updates. Portable bundles, recorded resource estimates, host-selection policies, and a fleet dashboard support distributed campaigns. An authenticated HTTPS gateway accepts submissions without fleet network access, while linked workflows connect calculation, validation, and rendering through checksummed transfers. Existing programs execute as separate processes, including vibe-qc; retrieved results can be inspected with vibe-view. We describe the architecture, operational limits, and chemistry workflow for the fixed 1.0.0 release. Its release-candidate verification record contains 13,149 passing tests and 37 skipped tests. These checks support documented software behavior; scientific correctness and suitability for a particular cluster require separate validation.

  • Open Access

    vibe-sketch for Chemical Structure Drawing and Calculation Preparation

    by Michael F. Peintinger

    Vibe-sketch connects chemical structure drawing, publication graphics and calculation preparation in an editable document. The browser and desktop interfaces combine a structure library with species identities, reactionroles,conformersandsavedcalculationrecipes;thesameoperationsareaccessiblethrough Python, command-line, REST and Model Context Protocol interfaces. We describe release 0.6.0 and the expanded template library in0.6.1,and demonstratetheworkflow for butadiene plus ethene forming cyclohexene.Twelvevibe-qc0.17.26calculationsconnectMACEoptimization,dispersion-freePBE optimization and Hessian evaluation, and restricted-reference correlated single points. Checks between calculationstagespreservethetransferredgeometriesand recordsofrejectedattempts.Thereported electronicreactionenergy is−190.54kJmol 1withintheexplicitlyreviewed restrictedprotocol;the - reference-wavefunction instabilities and the specific triples correction used remain documented. All 105 softwarechecks match theirexpected outcomes,including refusalofunsupported requests.Chemical schemesareexportedwithvibe-sketch0.6.0andresultorbitalsrenderedwithvibe-view3.0.0.The examplesdemonstratetraceablepreparationandreportingratherthangeneralchemicalaccuracyor exhaustive conformational sampling.

  • Open Access

    VIBE-VIEW and QVF: portable visualization and checksummed archives for quantum chemistry

    by Michael F. Peintinger

    Quantum-chemistry results often span separate files for structures, orbitals, scalar fields, and spectra. We present VIBE-VIEW, a viewer for molecular and periodic calculations, and QVF (Quantum Visualization Format), its producer-neutral archive format. QVF combines visualization data with provenance, citations, and optional job specifications and run records in one ZIP file. A JSON manifest declares payload formats and SHA-256 checksums, with dtypes and shapes for numeric arrays. Consumers verify payload integrity before use and explicitly report unsupported extensions. A published schema, extension registry, and twelve-archive conformance corpus support independent implementations. VIBE-VIEW provides renderers or panels for 41 of 42 canonical kinds, with bonds incorporated into structure rendering. Browser, desktop, terminal, headless, notebook, and Python interfaces support inspection and figure generation from the same archive, without requiring installation of the original producer. A water calculation demonstrates the workflow. VIBE-VIEW is licensed under MPL-2.0; the QVF specification and reference writer toolkit use Apache-2.0. Graphical TOC summary. QVF combines molecular and crystal structures, orbitals, spectra, and calculation provenance in one checksummed file. The VIBE-VIEW application opens these archives through interactive and scripted interfaces, supporting inspection and figure generation without installing the original calculation program.

  • Open Access

    VIBE-QC: an open-source, general-purpose electronic-structure code for molecules and solids developed with generative AI

    by Michael F. Peintinger

    We present VIBE-QC, a general-purpose open-source electronic-structure code for molecules and solids, developed almost entirely through generative-AI-assisted programming (“vibe-coding”) under expert human direction. The molecular stack provides Hartree–Fock and Kohn–Sham DFT over the full LIBXC library, analytic gradients and Hessians, Møller–Plesset theory and double hybrids, coupled cluster with local-correlation variants, multireference methods, excited states, dispersion, continuum solvation, effective core potentials, and semiempirical methods. The periodic stack treats one-, two-, and three-dimensional systems at multiple 𝒌 through three Fock-build routes. Production claims are validated by out-of-process cross-code comparisons against established reference programs, reaching the nanohartree scale for mean-field methods and 3.3 × 10−7 𝐸h for conventional CCSD(T); external parity for periodic total energies is open work at this release, tracked in the public issue tracker. VIBEQC adds a container-based results-storage and visualization format (QVF), an interactive viewer (VIBEVIEW), and a job queue (VIBE-QUEUE), under the MPL-2.0 license. Graphical TOC summary. VIBE-QC is an open-source, general-purpose electronic-structure code for molecules and solids that was built almost entirely through generative-AI-assisted programming under expert direction, and validated capability-by-capability where production status is claimed. It spans Hartree–Fock through coupled cluster, multireference, and excited states for molecules, and three periodic Fock-build routes for solids, and ships its own modern file format (QVF), viewer (VIBE-VIEW), and cluster-aware queue (VIBE-QUEUE).

  • Open Access

    VIBE-QC: an open-source, general-purpose electronic-structure code for molecules and solids developed with generative AI

    by Michael F. Peintinger

    We present VIBE-QC, a general-purpose open-source electronic-structure code for molecules and solids, developed almost entirely through generative-AI-assisted programming (“vibe-coding”) under expert human direction. The molecular stack provides Hartree–Fock and Kohn–Sham DFT over the full LIBXC library, analytic gradients and Hessians, Møller–Plesset theory and double hybrids, coupled cluster with local-correlation variants, multireference methods, excited states, dispersion, continuum solvation, effective core potentials, and semiempirical methods. The periodic stack treats one-, two-, and three-dimensional systems at multiple 𝒌 through three Fock-build routes. Production claims are validated by out-of-process cross-code comparisons against established reference programs, reaching the nanohartree scale for mean-field methods and 3.3 × 10−7 𝐸h for conventional CCSD(T); external parity for periodic total energies is open work at this release, tracked in the public issue tracker. VIBEQC adds a container-based results-storage and visualization format (QVF), an interactive viewer (VIBEVIEW), and a job queue (VIBE-QUEUE), under the MPL-2.0 license. Graphical TOC summary. VIBE-QC is an open-source, general-purpose electronic-structure code for molecules and solids that was built almost entirely through generative-AI-assisted programming under expert direction, and validated capability-by-capability where production status is claimed. It spans Hartree–Fock through coupled cluster, multireference, and excited states for molecules, and three periodic Fock-build routes for solids, and ships its own modern file format (QVF), viewer (VIBE-VIEW), and cluster-aware queue (VIBE-QUEUE).

  • Open Access

    Quantum-chemical study of stable, meta-stable and high-pressure alumina polymorphs and aluminum hydroxides

    by Michael F. Peintinger, Michael J. Kratz, Thomas Bredow

    The structure, properties and stability of 7 thermodynamically stable, meta-stable and high-pressure Al2O3 polymorphs and the structure and relative stability of 4 Al hydroxides were calculated with periodic hybrid density functional theory and compared with available experimental data.


  • The cyclic cluster model at hartree–fock level

    by Michael F. Peintinger, Thomas Bredow

    We present the implementation of the cyclic cluster model (CCM) formalism at the Hartree–Fock (HF) level. In contrast to other periodic models, the CCM is a Γ-point approach. Integration is carried out in real space within a finite interaction area determined by the size and the shape of the cluster that corresponds to a supercell of the solid, surface, or polymer. Particular care has to be taken for the proper treatment of three- and four-center integrals that involve basis functions located at the boundaries of the Wigner–Seitz supercell, which defines the interaction region. The similarity between the CCM formalism and molecular approaches allows in principle the application of sophisticated post HF methods to solid-state problems with only moderate modification of the molecular code. We show for selected model systems, that with our approach, the electronic structure and energetics obtained by the conventional supercell model is fully reproduced.