XtalPi is a technology company that integrates quantum physics–based computational methods, artificial intelligence, and robotic automation to accelerate and scale drug research and development. Operating across China and the United States, the company coordinates computational and experimental resources to meet the needs of clients and collaborators. Its iterative feedback loop between in silico (dry lab) modeling and wet-lab experimentation enables continuous optimization through cloud supercomputing and autonomous laboratory systems.

Foundation and Growth

Founded in 2015 in Boston, Massachusetts by three MIT postdoctoral physicists, XtalPi began with crystal structure prediction and solid-state analysis services. The accuracy of its crystal structure prediction (CSP) platform—validated in a blind test by Pfizer—led to collaborations in solid-form selection and polymorph screening, contributing to the solid-state development of nirmatrelvir (Paxlovid) during the COVID-19 pandemic. Between 2016 and 2024, XtalPi established an AI R&D center, launched the XtalPi Force Field (XFF) and Free Energy Perturbation (XFEP) solutions, opened wet labs for solid-state and synthesis research, developed autonomous laboratory prototypes, and expanded with a computational R&D center in Shenzhen and a pharmaceutical hub in Shanghai. It also created the XupremAb antibody discovery platform, deployed robotic smart network laboratories, and opened a demo lab in Boston. In 2024, the company expanded automation to chemistry, traditional Chinese medicine, and materials development and was listed on the Hong Kong Stock Exchange. XtalPi employs over 1,000 staff, holds 160+ granted patents, maintains 80+ partnerships across academia, pharma, and biotech, and operates 10,000+ m² of lab space.

Technology Platform

XtalPi’s hybrid AI + Quantum Mechanics (AI+QM) platform combines cloud-based GPU parallel computing with free energy perturbation (FEP) and force-field modeling to predict binding free energies, crystal structures, and drug-like properties. Experimental data from synthesis, screening, and characterization are fed back into active learning cycles for model refinement.

Key technologies include:

• XMolGen – generative chemistry platform enabling virtual screening, synthetic feasibility assessment, and candidate prioritization.
• XFEP – scalable FEP engine performing thousands of binding free energy calculations weekly.
• PatSight – NLP and optical extraction of chemical, activity, and ADMET data from patents and literature for SAR analysis.
• XFF – high-accuracy force-field computation system.
• XtalGazer – crystal structure prediction and polymorph analysis tool.
• ProteinGPT – proprietary large language model for protein design and biological sequence understanding.

XtalPi’s digital chemistry pipeline integrates intellectual property analysis, data fusion, generative molecular design, and rapid Design–Make–Test–Analyze (DMTA) loops through robotic synthesis and active learning. In structural biology, the company applies cryo-EM and micro-ED to elucidate binding modes, polymorphism relationships, and solid-state transitions.

Demonstrated achievements include:

• Identification of potent noncovalent GPX4 hits within 28 days.
• Discovery of flexible allosteric scaffolds using XMolGen + XFEP.
• Structural studies on clotrimazole, ketoprofen, and nirmatrelvir polymorphs.
• Mechanistic elucidation of SARM1 via cryo-EM.

Organization and Culture

The company is led by Shuhao Wen (Chairman & Co-founder), Jian Ma (CEO & Co-founder), and Lipeng Lai (Chief Innovation Officer & Co-founder). Its culture emphasizes market orientation, data-driven validation, and customer-focused process optimization. XtalPi has been recognized in Forbes China Enterprise Technology 50, Deloitte China Technology Fast 50, CB Insights Digital Health 150, MIT Technology Review China 50 Smartest Companies, and 36Kr AI Top 50.

Collaborations

XtalPi collaborates with major pharmaceutical and biotechnology companies in polymorph screening, solid-form selection, computational modeling, DEL/HTS assays, and structural characterization. Its software components are available through cloud or on-premise deployment, and its robotic wet-lab systems are adapted for chemical and materials R&D.

By connecting AI+QM computation with robotic experimentation, XtalPi institutionalizes a data-driven feedback cycle between dry and wet labs. Integrating patent mining, generative design, free energy computation, structural biology, and solid-state science, the company provides a unified R&D pipeline from target discovery to solid-form selection, aiming to enhance the speed, scale, and quality of drug development.