ATST-Tools
Governed, YAML-driven ASE transition-state toolkit (NEB / Dimer / Sella / IRC) for ABACUS & DeePMD.
GitHub →About
I work at the intersection of scientific discovery and engineering delivery in catalysis and materials simulation. My focus is on turning complex research problems into reproducible workflows and reusable open-source assets (from datasets and ML potentials to automation tooling). I care about shipping research-grade workflows that feel like production software — opinionated defaults, reproducibility, and good UX for scientists and AI agents.
Current interests: iron-based Fischer–Tropsch catalysis, surface reactions, active learning / fine-tuning for ML potentials, and community-facing tooling in the ABACUS / DeePMD ecosystem.
News
- 2026.06 DP-EVA (first author) published in Clean Energy (Oxford) — data-efficient fine-tuning of large atomistic models. DOI →
- 2026 ATST-Tools v2.2.0 released — governed ASE transition-state workflows for ABACUS & DeePMD. GitHub →
- 2025.08 Community: talks & recordings from ABACUS developer activities and a theoretical/computational chemistry meeting.
- 2025.03 FT2DP (first author) published in J. Mater. Informatics — large-atomic-model fine-tuning for iron-based Fischer–Tropsch synthesis. DOI →
Projects
DP‑EVA (dpeva)
Data-efficient active-learning fine-tuning of DPA universal potentials: uncertainty quantification, representative sampling, and automated DFT labeling.
GitHub →ABACUS user guide
Chinese documentation and onboarding for ABACUS open-source software.
GitHub →ABACUS toolchain
Cross-platform installation automation for ABACUS (major contributor, upstream).
GitHub →More projects on GitHub.
CV (brief)
Education
- Peking University (CCME) — PhD candidate, theoretical & computational chemistry (2022–present); advisor: Prof. Hong Jiang (TMC Group)
- East China University of Science and Technology (ECUST) — B.Sc. in Applied Chemistry (2018–2022); GPA 3.9/4.0, rank 1/270
Roles
- Project leader at Beijing Sidereus Intelligent Computing Technology Co., Ltd. (Sidereus AI), building the AsterFire Go scientific computing platform
- Open-source contributor / maintainer in the DeepModeling ecosystem (ABACUS / DeePMD)
- Community operator connecting developers and users
Selected publications
-
DP-EVA: Data-Efficient Fine-Tuning Framework via Maximizing Pre-Trained Knowledge of
Large Atomistic Models
Clean Energy (2026)
-
FT2DP: large atomic model fine-tuned machine learning potential for accelerating
atomistic simulation of iron-based Fischer–Tropsch synthesis
Journal of Materials Informatics (2025)
-
ABACUS: An electronic structure analysis package for the AI era
The Journal of Chemical Physics (2025)
-
DeePMD-kit v3: A Multiple-Backend Framework for Machine Learning Potentials
Journal of Chemical Theory and Computation (2025)
-
Switchable High-Valent Ag³⁺/Ag⁺ Redox Pair Stabilized in Polyoxometalate as Highly
Oxidative “Electron Shuttle” Catalysts
JACS Au (2025)
-
Efficient Catalytic Production of Hydrogen Peroxide Using Tin-containing Zeolite
Fixed Palladium Nanoparticles with Oxidation Resistance
Angewandte Chemie International Edition (2023)
-
Variable Lattice Monte Carlo: A Voronoi Tessellation-Guided Structural Exploration
Framework and Applications in Iron Carbides
The Journal of Physical Chemistry Letters (2026)
7 publications in total — full list on ORCID.
Contact
- Email quanmisaka [at] stu.pku.edu.cn
- GitHub https://github.com/QuantumMisaka
- ORCID 0009-0007-8286-9090
- ResearchGate Zhaoqing-Liu-8
- Bilibili 科学侠波本
- WeChat 公众号 波本的咖啡屋