The paper introduces KNIME (Konstanz Information Miner), a modular environment designed for easy visual assembly and interactive execution of data pipelines. KNIME is intended as a teaching, research, and collaboration platform, allowing users to integrate new algorithms, data manipulation, and visualization methods as modules. The architecture of KNIME is based on three main principles: a visual, interactive framework for combining data flows, modularity for easy distribution and independent development, and easy expandability through an open API and data abstraction framework. The paper details the data structures, node implementation, workflow management, and interactive views in KNIME. It also outlines the process of extending KNIME with new nodes and highlights ongoing developments, including meta-nodes, high-performance distributed computing, applications in life sciences, and web service integration.The paper introduces KNIME (Konstanz Information Miner), a modular environment designed for easy visual assembly and interactive execution of data pipelines. KNIME is intended as a teaching, research, and collaboration platform, allowing users to integrate new algorithms, data manipulation, and visualization methods as modules. The architecture of KNIME is based on three main principles: a visual, interactive framework for combining data flows, modularity for easy distribution and independent development, and easy expandability through an open API and data abstraction framework. The paper details the data structures, node implementation, workflow management, and interactive views in KNIME. It also outlines the process of extending KNIME with new nodes and highlights ongoing developments, including meta-nodes, high-performance distributed computing, applications in life sciences, and web service integration.