A Review of Automated Workflow Pipelines for Computational Chemists

Abstract Modern computational chemistry is a powerful tool for chemists to probe into material properties and to gain insight into the experimental results. In recent years, the development in artificial intelligence (AI) and machine learning (ML) has gained remarkable interest in computational chemistry. However, the accuracy of ML models highly depends on the fed data source. As a result, substantial high quality computational results from ab initio methods are required first to explore the potentials of AI and ML better. The extensive data demands from ML training lead to the appearance of high‐throughput quantum chemistry approach, where thousands of or tens of thousands of computation tasks are required. Batch processing of model creation and data processing by leveraging dedicated programs and codes is of significant importance to save the scientists from repeating laborious computer operations. This review focuses on the assistive tools and codes on automated workflows especially for high‐throughput quantum chemistry approaches.